EP1838891B1 - High strength sheet made from al-zn-cu-mg alloy with low internal stresses - Google Patents

Description

Field of the invention

This invention relates to a method for reducing the level of residual stresses throughout the thickness of 7xxx series aluminum alloy plates which are subjected to traction with permanent elongation.

State of the art

It is known that in aluminum alloys of the 7xxx series, ripening (natural aging) begins immediately after quenching. The underlying microstructural mechanism is related to the formation of Guinier-Preston zones by nucleation, and to the formation of metastable phases that precipitate from a supersaturated aluminum matrix. The nucleation and growth of these precipitates leads to a rapid increase of the elastic limit, because these precipitates hinder the dislocation displacement in the crystal lattice. The degree of hardening by these mechanisms at a given location in a thick plate will depend on the chemical composition, the quenching rate, the grain and subgrain structure of the metal, as well as its crystallographic texture.

Alloy plates of the 7xxx series (Al-Zn-Mg type alloys with or without copper) must be quenched quickly after being put in solution in order to be able to present, after income, high mechanical characteristics throughout their thickness. The presence at the time of quenching of strong thermal gradients near the surface of the strong plate leads to inhomogeneous plastic deformation. Therefore, when the sheet is completely cooled, it contains residual stresses (internal stresses). Specifically, there are compressive stresses near the surface, and tensile stresses in the center. The strength of these constraints depends on the alloy and the structure of the material, as well as the solution and quenching process; the order of magnitude is 200 MPa. A detailed description of the residual stresses in 7xxx alloys can be found in the following publications: JC Chevrier, F. Moreaux, G. Beck, J. Bouvaist, "Contribution to the study of thermal stress of quenching. Application to aluminum alloys. Scientific Memoirs - Revue de Metallurgie vol 72, n ° 1, p. 83-94 (1975) ); P. Jeanmart, J. Bouvaist, 7075 aluminum alloy, "Materials and Technology of Thermal Stresses in the Quenched Plate of High-Strength", Materials Science and Technology Vol. 1, no. 10, p. 765 - 769 (1985) ); D. Godard, Doctoral Thesis, National Polytechnic Institute of Lorraine, Nancy 1999, in particular pages 285 - 290 and 209 - 250 .

The most common methods for relaxing residual stresses in alloy plates of the 7xxx series use plastic deformation, either by L-direction tensile or by TC compression. The advantage of these methods is that they do not significantly affect the curing potential of the material during a subsequent stage of income. Traction is considered to be more effective than compression because it generally leads to a more homogeneous plastic deformation.

Licences US 6,159,315 and US 6,406,567 (Corus Aluminum Walzprodukte GmbH) discloses a method for relaxing the residual stresses of the hot plates after solution and quenching, which comprises a first step of L-shaped cold drawing, followed by a cold compression step in the TC direction.

The patent application WO 2004/053180 (Pechiney Rhenalu ) describes a method of relaxation of the residual stresses of a strong plate by compression on the fields. Even if it makes it possible to obtain plates with low residual energies, this method in compression is however difficult to implement.

Plastic deformation typically reduces the residual stresses by a factor of about 10. This is illustrated in FIG. figure 2 . Nevertheless, in practice, the residual stresses in thick semi-products considered to be identical can vary greatly. This may be related to the variation of their chemical composition, but also and especially to the variation of the parameters of the manufacturing processes, such as casting, rolling, quenching, traction and income; the influence of these process parameters on the level of residual stresses in the finished product is not always well understood. Some process modifications actually lead to a reduction in the level of residual stresses (such as the choice of a slower quenching or a higher tempering temperature), but they also modify the tradeoff between certain properties which are important for structural applications, such as, typically, mechanical strength, damage tolerance and corrosion resistance. This state of the art is known from the following publications: R. Habachou, M. Boivin, "Numerical predictions of quenching and releasing by stretching of aluminum alloy cylindrical bars", Journal of Theoretical and Applied Mechanics, Vol 4, pp. 701-723, 1985 ; JC Boyer and M. Boivin, "Numerical calculations of residual stress relaxation in quenched plates", Materials Science and Technology Vol. 1985 pp. 786-792 ; R. Vignaud, P.Jeanmart, J. Bouvaist, B. Dubost (1990), "Detensioning by plastic deformation", Physics and mechanics of metal shaping, Oleron Summer School, directed by F. Moussy and P. Franciosi, published in CNRS Press, 1990, pp. 632-642 .

The critical influence of residual stresses on the distortion during machining has been widely described in the literature. In the aerospace industry, complex parts are often machined from thick aluminum alloy sheets; this often leads to more than 80% chips. Too much distortion in machining must be compensated for by complex and costly corrective measures, such as: (a) mechanical straightening, (b) shot peening, (c) optimization of localization of the target part in the thickness of the plate, that is to say with respect to the depth profile of the residual stresses, or (d) the modification of the shape of the part in order to minimize its deformation ( it being understood that the permanent deformation of the machined part is small if its shape is close to a symmetrical shape with respect to the longitudinal axis of the thick plate in which said part is machined). Aircraft manufacturers therefore prefer heavy plates whose residual stresses are not only lower, but also under control, that is to say with a small variation for a given type of product (alloy, thickness, metallurgical state). ).

Licences EP 0 731 185 and US 6,077,363 describe a method to reduce residual stresses in 2024 alloy plate. The optimization of the manganese content and the outlet temperature of the hot rolling mill to obtain a recrystallization rate of more than 50% throughout the thickness. Such a sheet shows a better homogeneity of the mechanical characteristics as a function of the thickness, as well as a reduced level of residual stresses after traction.

EP1231290 describes the manufacture of wrought products by rolling, spinning or forging aluminum alloy type AlZnMgCu high mechanical strength, used in particular in aircraft construction, in particular for the extrados of aircraft wings. 7449-thick alloy sheets 38 mm thick were made. The composition of the alloy was as follows (% by weight): Zn = 8.11 Mg = 2.19 Cu = 1.94 Si = 0.04 Fe = 0.07 Zr = 0.09 Cr = 0.005 Ti = 0.025 remains aluminum and impurities (<0.05 each). The sheets were pre-expanded to change from a plate width of 1100 mm to 2500 mm to a hot rolling up to 38 mm with an outlet temperature of 378 ° C solution at 475 ° C, a quenched with cold water, and a controlled pull at 2.8% permanent elongation after a delay of 1 h after quenching.

For 7xxx heavy plates, it is generally preferred to keep a largely unrequistallised microstructure, especially for applications that require high toughness, such as structural elements for aircraft. This is apparent from the publication of F. Heymes, B. Commet, B. Dubost, P. Lassince, P. Lequeu, and GM Raynaud, "Development of new Al alloys for distortion free machined aluminum aircraft components", published in 1st International Non Ferrous Processing and Technology Conference , St. Louis, Missouri, 1997, 249-255 .

Residual stresses in thick plates can be determined by the method of successive machining described in the publication of Heymes, Commet et al., Referenced above. A method based on this publication is described in detail below.

The aim of the present invention is to present a method for obtaining thick aluminum alloy plates of the 7xxx series which, in the tractionned state, in the matured state or in any state of artificial aging, exhibits a lower residual stress level, without degrading mechanical strength and damage tolerance. More particularly, it is desired to have plates that do not deform during machining, which is observed when the total elastic energy stored in the sheet, W, is less than 2 kJ / m ³ and preferably less than 1 kJ / m ³ .

Object of the invention

The subject of the invention is a method for manufacturing thick plates made of Al-Zn-Cu-Mg type alloy comprising between 4 and 12% of zinc, less than 4% of magnesium and less than 4% of copper, minor elements ≤ 0.5% each, the rest aluminum, said process comprising hot rolling, dissolution, quenching, traction controlled with a permanent elongation greater than 0.5%, as well as aging,
characterized in that delay D between the end of quenching and the beginning of the controlled pull is less than 2 hours, and preferably less than 1 hour.

Another object of the present invention is a thick plate alloy type AI-Zn-Cu-Mg comprising between 4 and 12% zinc, less than 4% magnesium and less than 4% copper, minor elements ≤ 0, 5% each, the rest aluminum, hot rolled, dissolved, quenched, tractionned with a permanent elongation greater than 0.5%, aged, characterized in that its total elastic energy is less than or equal to $$\mathrm{W}\left[\mathrm{K\; J}/{\mathrm{<figure-callout\; id="3"\; label="m"\; filenames="imgf0007.png"\; state="\{\{state\}\}">m</figure-callout>}}^{\mathrm{3}}\right]=\mathrm{0},\mathrm{54}+\mathrm{0},\mathrm{013}\left({\mathrm{R}}_{\mathrm{p}\mathrm{0},\mathrm{2}\left(\mathrm{The}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)\mathrm{.}$$

Yet another subject of the invention is a control batch or a heat treatment batch of Al-Zn-Cu-Mg alloy thick plates comprising between 4 and 12% of zinc, less than 4% of magnesium and less of 4% copper, minor elements ≤ 0.5% each, the rest aluminum, in the dissolved state, quenched, triturated and aged, characterized in that the total elastic energy W (expressed in kJ / m ³ ) plate shows a standard deviation less than or equal to $$\mathrm{0},\mathrm{20}+\mathrm{0},\mathrm{0030}\left({\mathrm{P}}_{\mathrm{p}\mathrm{0},\mathrm{2}\left(\mathrm{The}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)\mathrm{around\; an\; average\; value}\mathrm{.}$$

Description of figures

• The figure 1 shows schematically the definition of the three main directions in a sheet.
• The figure 2 shows schematically a traction curve. Curve 2 represents the state of stresses in the core of the sheet. Curve 1 shows the state of surface stresses. This figure shows the principle of controlled tensile stress relief: before the controlled tension, the stress deviation between the surface and the core is defined by x and -x. Controlled traction reduces this difference (defined by y and y) typically by a factor of 10.
• The figure 3 shows the definition of the parameters h, l and w of a sheet. Below, we see schematically the strain gauge (with its connecting wire).
• The figure 4 schematically shows the measurement and calculation sequences to determine a residual stress profile in the sheet thickness using the successive layer removal method.
• The figure 5 shows schematically the critical part of the method according to the invention. D denotes the time interval between the end of quenching and the beginning of the controlled pull.
• The figure 6 shows the kinetics of maturation of thick plates in alloys 7010 and 7050 for two different quenching speeds. The abscissa shows the limit of elasticity in the direction L, the ordinate the time of maturation.
• The figure 7 shows the effect of increasing the variation of elastic limit values on the residual stress profiles after quenching.
• The figure 8 shows the total elastic energy as a function of the thickness for batches of alloy sheets 7xxx according to the invention (with D ≤ 1 hour) (open points) and according to the state of the art (with D ≥ 8 hours) (black squares).

Description of the invention a) Terminology

Unless stated otherwise, all the information relating to the chemical composition of the alloys is expressed in percent by weight. The designation of the alloys follows the rules of The Aluminum Association, known to those skilled in the art. The term "Al-Zn-Cu-Mg alloy" refers to an aluminum-based alloy that contains the elements of zinc, copper and magnesium alloy; such an alloy may contain in addition to other alloying elements as well as other elements, the presence of which may be intentional or not, for example impurities.

The metallurgical states are defined in the European standard EN 515. The chemical composition of standardized aluminum alloys is defined for example in the standard EN 573-3. Unless otherwise stated, the static mechanical characteristics, ie the breaking strength R _m , the yield _stress R _p0,2 , and the elongation at break A, are determined by a tensile test according to EN 10002-1 standard, the location and direction of specimen collection being defined in EN 485-1. K _IC toughness was measured according to ASTM E 399.

Unless otherwise stated, the definitions of the European standard EN 12258-1 apply.

In the context of the present invention, a "thick sheet" designates a sheet whose thickness is greater than 6 mm.

The term "inspection lot" is defined in EN 12258-1; it means an expedition or part of an expedition, subject to control, and which includes products of the same quality or alloy, of the same shape, metallurgical condition, size, geometry, thickness or cross-section, and which have been produced by the same processes.

The term "heat treatment batch" means a quantity of products of the same quality or alloy, of the same shape, thickness or cross-section, and which have been produced in the same way, including the heat treatment or solution solution followed quenching was carried out in a single charge; several batches can be dissolved in the same heat treatment batch.

The term "aging" includes natural aging at room temperature (also called "ripening"), as well as any artificial aging (also known as "income").

The term "machining" includes any material removal process such as turning, machining, milling, drilling, reaming, tapping, EDM, grinding, polishing, chemical machining.

The term "structural element" refers to an element used in mechanical engineering for which the static and / or dynamic mechanical characteristics are of particular importance for the performance and integrity of the structure, and for which a calculation of the structure is usually prescribed or performed. It is typically a mechanical part whose failure is likely to endanger the safety of said construction, its users, its users or others. For an aircraft, these structural elements include the elements that make up the fuselage (such as fuselage skin (fuselage skin in English), stiffeners or stringers, bulkheads, fuselage (circumferential frames), the wings (such as the wing skin), the stiffeners (stringers or stiffeners), the ribs (ribs) and spars) and the empennage composed in particular of stabilizers Horizontal and vertical (horizontal or vertical stabilizers), as well as floor beams, seat rails and doors.

The term "monolithic structure element" refers to a structural element that has been obtained, most often by machining, from a single piece of laminated, extruded, forged or molded semi-finished product, such as riveting, welding, gluing, with another piece.

The directions L (Long direction), TL (cross-long direction) and TC (short-path direction) in a rolled product refer to the rolling direction corresponding to the direction L. These three directions are defined on the figure 1 .

b) Determination of residual stresses

In the context of the present invention, the residual stresses were determined using the method based on the successive removal of layers described in the publication " Development of New Alloy for Distortion Free Machined Aluminum Aircraft Components ", F. Heymes, B.Commet, B.Dubost, P.Lassince, P.Lequeu, GM.Raynaud, in 1st International Non-Ferrous Processing & Technology Conference, 10- March 12, 1997 - Adams's Mark Hotel, St. Louis, Missouri .

This method is especially applicable to tractionned heavy plates, in which the state of stress can be considered as biaxial; the two main components being located in the directions L and TL, and there is therefore no component in the direction TC. This method is based on the determination of the residual stresses in the directions L and TL on rectangular bars, cut in full thickness of the sheet in the direction parallel to the indicated directions. These bars are machined in the TC step by step direction. After each step, the stress and / or deflection are measured and the thickness of the bar is measured. A particularly preferred method is to set a strain gauge at mid-length of the bar, on the surface opposite to that which is machined. This makes it possible to calculate the residual stress profiles in the L and TL directions. The bar should be long enough to avoid edge effects. The recommended dimensions according to the thickness of the sheet are shown in Table 1. Table 1 Dimensions [mm] used for the method of successive layer removal Thickness of the sheet (h) Width (w) Length (1) 20 <h≤100 24 ± 1 5h ± 1 h> 100 30 ± 1 5h ± 1

Unidirectional strain gauges with compensation for thermal expansion are glued to the lower surface of the bar (see figure 3 ), following the manufacturer's instructions. Then they are covered with an insulating lacquer. The value read on each of these gauges is taken as zero.

A measurement is taken after each machining pass. Typically, between 18 and 25 passes are used to obtain a sufficient number of points to calculate the stress profile. The machining depth must not be less than 1 mm, in order to obtain a good quality of cut; for very thick sheets, it can reach 10 mm. Chemical machining can also be used to remove a very small thickness of metal. The machining pitch should be the same for both samples (i.e. in the L direction and in the TL direction).

After each machining pass, the bar is detached from the vice, and the temperature is allowed to stabilize before measuring the deformation. At each step i , the thickness h (i) and the strain ε (i) are noted . The scheme of the figure 4 shows how we collect this data.

• Pour i = 1 à N-1 : $$u\left(i\right)=-E\frac{\left(\epsilon \left(i+1\right)-\epsilon \left(i\right)\right)h{\left(i+1\right)}^2}{\left[h\left(i\right)-h\left(i+1\right)\right]\left[3h\left(i\right)-h\left(i+1\right)\right]}-S\left(i\right)$$
  
  avec : $$S\left(i\right)=E{\displaystyle \sum _{k=1}^{i-1}}\left(\epsilon \left(k+1\right)-\epsilon \left(k\right)\right)\left[1-\frac{3h\left(k\right)\left(h\left(i\right)+h\left(i+1\right)\right)}{\left(3h\left(k\right)-h\left(k+1\right)\right)h\left(k+1\right)}\right]$$
  
  où E est le module de Young de la tôle épaisse. On obtient ainsi deux profiles : u(i)_L et u(i)_LT qui correspondent à des barres à section rectangulaire dans les directions L et TL. Les profiles de contraintes dans la tôle sont obtenus par les équations suivantes :
• Pour i = 1 à N-1 $$\sigma {\left(i\right)}_L=\frac{u{\left(i\right)}_L+\mathit{vu}\left(i\right)\mathit{LT}}{1-{\mathrm{<figure-callout\; id="2"\; label="\nu "\; filenames="imgf0002.png,imgf0004.png"\; state="\{\{state\}\}">\nu </figure-callout>}}^2}$$
  
  $$\sigma {\left(i\right)}_{\mathit{LT}}=\frac{u{\left(i\right)}_{\mathit{LT}}+\mathit{vu}{\left(i\right)}_L}{1-{\mathrm{<figure-callout\; id="2"\; label="\nu "\; filenames="imgf0002.png,imgf0004.png"\; state="\{\{state\}\}">\nu </figure-callout>}}^2}$$
  
  où v est le coefficient de Poisson de la tôle forte. On peut ensuite calculer l'énergie stockée dans la tôle (W_L, W_LT et W) à partir des equations : $$W_L=\frac{500}{\mathit{Eh}}{\displaystyle \sum _{i=1}^{N-1}}\sigma {\left(i\right)}_L\left[\sigma {\left(i\right)}_L-\mathit{\nu \sigma }{\left(i\right)}_L\right]\mathit{dh}\left(i\right)$$
  
  $$W_{\mathit{LT}}=\frac{500}{\mathit{Eh}}{\displaystyle \sum _{i=1}^{N-1}}\sigma {\left(i\right)}_{\mathit{LT}}\left[\sigma {\left(i\right)}_{\mathit{LT}}-\mathit{\nu \sigma }{\left(i\right)}_L\right]\mathit{dh}\left(i\right)$$
  
  $$W=W_L+W_{\mathit{LT}}$$
  
  où W_L représente l'énergie élastique stockée qui résulte du profil de contraintes résiduelles dans la direction L, et W_LT représente l'énergie stockée qui résulte du profil de contraintes résiduelles dans la direction TL. W est l'énergie élastique totale dans la tôle (exprimé en kJ ou kJ/m³). La méthode de mesure des contraintes et d'obtention des énergies élastiques stockées est décrite ci-dessus d'une manière précise en donnant par exemple les dimensions des barreaux qui sont utilisés en pratique. Il faut noter que ces dimensions ne sont pas obligatoires et ne limitent pas la méthode. La largeur du barreau n'a pas d'influence sur le résultat. Une longueur de deux fois h plus trois fois la longueur de la jauge est suffisante dans le cas de mesures à l'aide de jauges de déformation. Les dimensions données sont issues de l'expérience pratique et ont été adaptées aux moyens d'usinage et de mesures utilisés. L'homme du métier sera aisément en mesure de sélectionner d'autres dimensions sans altérer les résultats.

These data make it possible to calculate the initial stress profile in each bar in the form of a curve u (i), which corresponds to the average stress in the layer removed during the machining step i , given by the following equations:

• For i = 1 to N-1: $$u\left(i\right)=-E\frac{\left(\epsilon \left(i+1\right)-\epsilon \left(i\right)\right)h{\left(i+1\right)}^2}{\left[h\left(i\right)-h\left(i+1\right)\right]\left[3h\left(i\right)-h\left(i+1\right)\right]}-S\left(i\right)$$
  
  with: $$S\left(i\right)=E{\displaystyle \underset{k=1}{\overset{i-1}{\Sigma }}}\left(\epsilon \left(k+1\right)-\epsilon \left(k\right)\right)\left[1-\frac{3h\left(k\right)\left(h\left(i\right)+h\left(i+1\right)\right)}{\left(3h\left(k\right)-h\left(k+1\right)\right)h\left(k+1\right)}\right]$$
  
  where E is the Young's modulus of the thick plate. Two profiles are thus obtained: u (i) _L and u (i) _LT which correspond to rectangular section bars in the L and TL directions. The stress profiles in the sheet are obtained by the following equations:
• For i = 1 to N-1 $$\sigma {\left(i\right)}_{\mathrm{The}}=\frac{u{\left(i\right)}_{\mathrm{The}}+\mathit{seen}\left(i\right)\mathit{LT}}{1-{\mathrm{<figure-callout\; id="2"\; label="\nu "\; filenames="imgf0002.png,imgf0004.png"\; state="\{\{state\}\}">\nu </figure-callout>}}^2}$$
  
  $$\sigma {\left(i\right)}_{\mathit{LT}}=\frac{u{\left(i\right)}_{\mathit{LT}}+\mathit{seen}{\left(i\right)}_{\mathrm{The}}}{1-{\mathrm{<figure-callout\; id="2"\; label="\nu "\; filenames="imgf0002.png,imgf0004.png"\; state="\{\{state\}\}">\nu </figure-callout>}}^2}$$
  
  where v is the Poisson's ratio of the strong plate. The energy stored in the sheet (W _L , W _LT and W) can then be calculated from the equations: $$W_{\mathrm{The}}=\frac{500}{\mathit{Well}}{\displaystyle \underset{i=1}{\overset{\mathrm{NOT}-1}{\Sigma }}}\sigma {\left(i\right)}_{\mathrm{The}}\left[\sigma {\left(i\right)}_{\mathrm{The}}-\mathit{\nu \sigma }{\left(i\right)}_{\mathrm{The}}\right]\mathit{dh}\left(i\right)$$
  
  $$W_{\mathit{LT}}=\frac{500}{\mathit{Well}}{\displaystyle \underset{i=1}{\overset{\mathrm{NOT}-1}{\Sigma }}}\sigma {\left(i\right)}_{\mathit{LT}}\left[\sigma {\left(i\right)}_{\mathit{LT}}-\mathit{\nu \sigma }{\left(i\right)}_{\mathrm{The}}\right]\mathit{dh}\left(i\right)$$
  
  $$W=W_{\mathrm{The}}+W_{\mathit{LT}}$$
  
  where W _L represents the stored elastic energy that results from the residual stress profile in the L direction, and W _LT represents the stored energy that results from the residual stress profile in the TL direction. W is the total elastic energy in the sheet (expressed in kJ or kJ / m ³ ). The method for measuring the stresses and obtaining the stored elastic energies is described above in a precise manner, for example by giving the dimensions of the bars that are used in practice. It should be noted that these dimensions are not mandatory and do not limit the method. The width of the bar has no influence on the result. A length of two times plus three times the length of the gauge is sufficient in the case of measurements using strain gauges. The given dimensions are derived from practical experience and have been adapted to the machining and measuring means used. Those skilled in the art will readily be able to select other dimensions without altering the results.

Similarly, other techniques can be used to measure the stress gradient in the thickness of the sheets. After obtaining the stress profiles σ _L and σ _LT in the thickness, the same formulas of the incremental sums above make it possible to calculate the stored energies W _L and W _LT . It is therefore possible to obtain the stored energies by any technique allowing measurements of stresses in the thickness.

c) Detailed description of the invention

• 4 < Zn < 12 ; Mg < 4 ; Cu < 4 ;
• éléments mineurs ≤ 0,5 chacun
• le reste aluminium,

et qui sont traités par mise en solution, trempe et traction contrôlée.The present invention applies to plates, and especially to plates, made of aluminum alloy of the 7xxx series, the chemical composition of which satisfies the following criteria:

• <Zn <12; Mg <4; Cu <4;
• minor elements ≤ 0.5 each
• the rest aluminum,

and which are treated by dissolution, quenching and controlled pulling.

According to the invention, the problem is solved by a modification of the manufacturing process so that the ripening (natural aging) between the end of the quenching and the beginning of the controlled pull is minimized so that the energy total elasticity (W) in the return state remains below a certain limit value. This limit value represents a maximum value to keep the machining deformity at a level that is still acceptable; for most applications, this limit value is 2 kJ / m ³ for a sheet having a thickness of between 60 mm and 100 mm, and preferably of 1.5 kJ / m ³ . For particularly complex parts, it must be 1 kJ / m ³ .

The figure 5 shows the diagram of the heat treatment process that a sheet undergoes after rolling. The dissolution can be carried out in a single stage, in several stages, or in ramp with or without definite stage. The same is true of income. The critical phase in the context of the present invention is the delay D between the end of the quenching and the beginning of the controlled pull. The inventors have found that a long delay D leads to a greater heterogeneity of the mechanical characteristics between areas near the surface and areas near the mid-thickness of the material. This heterogeneity can be mainly attributed to differences in cooling rate in the thickness of the sheet. The figure 6 shows the evolution of the yield strength L, determined close to the surface and at mid-thickness, as a function of the curing time for very high alloy plates AA7010 and AA7050 and for different nominal quenching rates. These quenching speeds were obtained on tensile test pieces but they are representative of the differences in quenching velocity observed between the surface and the core of a thick sheet. It can be seen that the difference between the levels of mechanical strength increases over time.

The inventors have found that the variation of the residual stresses across the thickness of the alloy sheets 7xxx depends on (i) the variation of the cooling rates and the plastic deformation during quenching, (ii) the heterogeneities of the microstructure granular structure and texture that are generated during rolling, and (iii) local variations in chemical composition that result from the casting process (including solidification and homogenization). Between the end of the quenching and the beginning of the traction, a maturation is observed throughout the thickness of the sheet, but the speed of this maturation depends on the thickness: the limit of elasticity increases faster close to a surface only half thickness. This is probably due to the kinetics of precipitation: on the one hand, the content of the solid solution supersaturated in potentially hardening elements is greater close to the surface than in mid-thickness (because the adopted semi-continuous casting process leads a macro-segregation such that the concentration of eutectic elements, such as Cu, Mn and Zn, is higher close to the surface, and the cooling rate during casting is also greater), and on the other hand, a greater density of heterogeneous sites (vacancies, dislocations, etc.) is found close to the surface, which facilitates precipitation and which results from the greater rate of cooling and greater plasticity during quenching.

The inventors have found by a calculation based on a finite element model that an increase in the heterogeneity of the mechanical characteristics (i.e. the elastic limit or the coefficients of work hardening) leads to an increase in the residual stresses after pulling. The figure 7 shows the effect of increasing the variation of elastic limit values on the residual stress profiles after quenching.

This attempt at metallurgical explanation of the process according to the invention does not, however, imply any limitation of the present invention to the underlying phenomena. Moreover, the inventors have found that the effect is greater in reality than the values obtained by the mathematical model.

Finally, a change in the manufacturing process that would improve the homogeneity of the flow limits (R _{pO 2} ) in the thickness of the strong plate after quenching, would result in a reduction of the residual stresses after controlled pulling or after any stress relief. plastic deformation.

The method according to the invention does not give an improved result in the case of other alloys with structural hardening, such as the 2xxx and 6xxx series alloys. For very heavy alloys, that is to say having a content of Zn> 12%, Mg> 4% and Cu> 4%, the stored energy is very high, and the improvement obtained with the process according to the invention. invention does not appear to be significant. These alloys also have difficulty responding to solution treatment.

Dans cette équation, R_p0,2(L) désigne la limite élastique de la tôle finie mesurée selon les normes EN 10002-1 et EN 485-1. L'influence de l'épaisseur sur le niveau de contraintes résiduelles et l'énergie élastique totale est ici exprimée en termes de la limité d'élasticité, mesurée comme préconisé par la norme EN 485-1. Le procédé selon l'invention peut être appliqué avantageusement à la fabrication d'une pluralité de tôles dont l'épaisseur se situe entre environ 10 mm et environ 250 mm, et encore plus avantageusement à des tôles dont l'épaisseur est supérieure à 25 mm, mais ces valeurs ne sont pas limitatives.The method according to the invention makes it possible to manufacture sheets characterized by a value of the total elastic energy which is less than or equal to $$\mathrm{W}\left[\mathrm{K\; J}/{\mathrm{<figure-callout\; id="3"\; label="m"\; filenames="imgf0007.png"\; state="\{\{state\}\}">m</figure-callout>}}^{\mathrm{3}}\right]=\mathrm{0},\mathrm{54}+\mathrm{0},\mathrm{013}\left({\mathrm{R}}_{\mathrm{p}\mathrm{0},\mathrm{2}\left(\mathrm{The}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)\mathrm{.}$$

In this equation, R _{p0,2 (L)} denotes the yield strength of the finished sheet measured according to EN 10002-1 and EN 485-1. The influence of the thickness on the level of residual stresses and the total elastic energy is here expressed in terms of the limited elasticity, measured as recommended by the EN 485-1 standard. The method according to the invention can advantageously be applied to the manufacture of a plurality of sheets whose thickness is between about 10 mm and about 250 mm, and even more advantageously to sheets whose thickness is greater than 25 mm. , but these values are not limiting.

et de manière préférée inférieur ou égal à $$\mathrm{0},\mathrm{20}+\mathrm{0},\mathrm{0030}\left({\mathrm{P}}_{\mathrm{p}\mathrm{0},\mathrm{2}\left(\mathrm{L}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)\mathrm{.}$$

Dans cette equation, R_p0,2(L) désigne la moyenne des mesures de R_p0,2(L) effectuées selon la norme pour chacune des tôle finies du lot, selon les normes EN10002-1 et EN485-1.The method according to the invention also makes it possible to reduce the dispersion between the values of W for a plurality of sheets belonging to the same control lot or batch of heat treatment, so that all the plates have a standard deviation of the total elastic energy W of the various sheets around a mean value less than or equal to $$\mathrm{0},\mathrm{20}+\mathrm{0},\mathrm{0086}\left({\mathrm{P}}_{\mathrm{p}\mathrm{0},\mathrm{2}\left(\mathrm{The}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)$$

and preferably less than or equal to $$\mathrm{0},\mathrm{20}+\mathrm{0},\mathrm{0030}\left({\mathrm{P}}_{\mathrm{p}\mathrm{0},\mathrm{2}\left(\mathrm{The}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)\mathrm{.}$$

In this equation, R _{p0,2 (L)} denotes the average of the measurements of R _{p0,2 (L)} carried out according to the standard for each of the finished sheets of the batch, according to the standards EN10002-1 and EN485-1.

The standard deviation between the measurements of the total elastic energy W of the different sheets of a batch can depend on the number of sheets contained in the batch. In particular, a standard deviation obtained on two measurements is weakly significant and can randomly be very high or very low. From 3 sheets, the standard deviation of the measurements can be considered but, in a preferred manner, the control or heat treatment batches used in the context of the present invention contain at least 5 sheets.

The use of the method according to the invention enables the manufacturer to guarantee that such a control batch or such a batch of heat treatment comprises sheets whose average total elastic energy is less than 3 kJ / m ³ . Preferably, this average value is less than 2 kJ / m ³ , and a value less than 1 kJ / m ³ is preferred, which requires excellent control of the critical processes and very rigorous management of the product streams at the stages of production. dissolution, quenching and traction. Indeed, the implementation of the method according to the invention may require an adaptation of the metal flows inside the plant, because if the producer wants to produce plates with a delay D less than a few hours, it is necessary to synchronize the tempering furnace with the traction bench. In practice, this means minimizing the intermediate stock between these two machines; this applies in particular to particularly preferred embodiments with D <1 hour or D <30 minutes. The patent application EP 1 231 290 A1 described in Example 1 a 7449 thick alloy sheet 38 mm for which the controlled pull was performed 1h after quenching; this document however does not give any teaching on the interest of this short duration. The method according to the invention made it possible to produce control batches or heat treatment batches for which the delay D between the end of the quenching and the beginning of the controlled pull is systematically less than 2 hours, which made it possible to minimize the average and the standard deviation of the total elastic energy W of the sheets of these batches. The industrial manufacture of such a control lot, however, requires a reorganization of product flows around the machines necessary for the implementation of the method according to the invention.

In another embodiment of the invention, the maturation is carried out at low temperature, that is to say at a temperature below 10 ° C. and preferably at a temperature below 5 ° C., which makes it possible to obtain similar results in terms of total elastic energy W for delays D between 2h and 3h.

Other preferred embodiments of the invention are indicated in the dependent claims. The invention is particularly advantageous for thick plates of AA7010, 7050, 7056, 7449, 7075, 7475, 7150, 7175 alloys.

The advantage of the process according to the invention is the overall reduction of the stress level in the heavy plates. This generally reduces the deformation during machining.

Another advantage of the process according to the invention is that the control of the time which elapses between the end of the quenching and the beginning of the traction also makes it possible to reduce the dispersion of the stress level which is observed between different sheets nominally. identical, even within the same manufacturing batch or batch of heat treatment. This allows a better standardization of the machining processes for a given series of products, and reduces the number of incidents during the manufacture of machined parts in the machine shop.

In the examples which follow, advantageous embodiments of the invention are illustrated by way of illustration. These examples are not limiting in nature.

Examples Example 1

Three AA7010 alloy rolling plates were cast by semi-continuous casting. After homogenization, they were hot-rolled to a thickness of 100 mm. At the outlet of the hot rolling mill, they were subjected to quenching followed by a controlled pull, and finally to a treatment of income. The metallurgical state of the three products A1, A2 and A3 thus obtained was the T7651 state. For these three products, all manufacturing parameters were nominally identical and well controlled. The only difference was the waiting time D between the end of quenching and the beginning of tensile stress relief.

By a similar process, three AA7050 alloy rolling plates were converted by homogenization, hot rolling to a thickness of 100 mm, quenching, controlled pulling and tempering. The metallurgical state of the three products B1, B2 and B3 thus obtained was the T7451 state. For all three products, all manufacturing parameters were nominally identical and well controlled, and the only difference was the waiting time D between the end of quenching and the beginning of tensile stress relief.

Table 2 shows the stored elastic energy of the various sheets obtained, determined in the final state. When reducing the waiting time D between the end of the quenching and the beginning of tensile stress relief, a reduction in the overall stress level as measured by W _L , W _LT and W is observed. Table 2 Stored elastic energy (in the final state) as a function of the ripening time for three alloy plates 7010 and 7050. sheet metal Alloy / state Maturation time D [h] W [kJ / m ³ ] W _L [kJ / m ³ ] W _LT [kJ / m ³ ] A1 7010 T7651 1.17 1.02 0.8 0.22 A2 7010 T7651 9 1.76 1.37 0.4 A3 7010 T7651 48.92 2.37 1.74 0.63 B1 7050 T7451 1.25 1.22 0.84 0.38 B2 7050 T7451 8.83 2.28 1.57 0.71 B3 7050 T7451 49.08 3.15 2.02 1.12

The static mechanical characteristics were measured in the state of final heat treatment in directions L, TL and TC at ¼, ½ and ¾ thickness. The results are shown in Tables 3, 4 and 5. It can be seen that the maturation time D has no significant influence on the static mechanical characteristics. Table 3: Static mechanical characteristics (L direction) in the final state as a function of the maturation time D for alloy plates 7010 and 7050 sheet metal Alloy / state Maturation time D [h] Location R _{m (L)} [MPa] R _{p0.2 (L)} [MPa] A _(L) [%] A1 7010 T7651 1.17 ¼ thickness 524 479 14.0 ½ thickness 519 468 12.7 ¾ thickness 533 471 11.0 A2 7010 T7651 9 ¼ thickness 529 480 14.4 ½ thickness 523 477 11.5 ¾ thickness 539 480 9.6 A3 7010 T7651 48.92 ¼ thickness 521 472 12.6 ½ thickness 516 466 9.2 ¾ thickness 528 472 8.2 B1 7050 T7451 1.25 ¼ thickness 536 482 13.0 ½ thickness 519 465 10.4 ¾ thickness 531 470 9.6 B2 7050 T7451 8.83 ¼ thickness 534 479 14.2 ½ thickness 519 461 10.8 ¾ thickness 533 469 8.7 B3 7050 T7451 49.08 ¼ thickness 534 478 14.2 ½ thickness 519 459 10.5 ¾ thickness 531 463 9.4 Static mechanical characteristics (TL direction) in the final state as a function of the aging time D for alloy plates 7010 and 7050 sheet metal Alloy / state Maturation time D [h] Location R _{m (TL)} [MPa] R _{p0.2 (TL)} [MPa] A _(TL) [%] A1 7010 1.17 ¼ thickness 529 470 10.4 T7651 ½ thickness 527 464 9.4 ¾ thickness 513 446 9.2 A2 7010 9 ¼ thickness 536 475 11.0 T7651 ½ thickness 534 478 8.4 ¾ thickness 521 463 8.1 A3 7010 48.92 ¼ thickness 527 461 10.1 T7651 ½ thickness 526 463 7.8 ¾ thickness 511 452 8.0 B1 7050 1.25 ¼ thickness 541 461 10.6 T7451 ½ thickness 526 456 6.6 ¾ thickness 516 443 6.7 B2 7050 8.83 ¼ thickness 541 464 9.6 T7451 ½ thickness 528 464 6.9 ¾ thickness 519 447 7.2 B3 7050 49.08 ¼ thickness 538 467 10.8 T7451 ½ thickness 527 451 7.8 ¾ thickness 513 440 6.4 Static mechanical characteristics (TC direction) in the final state as a function of the maturation time D for alloy plates 7010 and 7050 sheet metal Alloy / state Maturation time D [h] Location R _{m (TC)} [MPa] R _{p0.2 (TC)} [MPa] A _(TC) [%] A1 7010 T7651 1.17 ¼ thickness 517 449 6.5 ½ thickness 508 432 7.7 ¾ thickness 518 455 6.3 A2 7010 T7651 9 ¼ thickness 521 455 5.7 ¼ thickness 520 438 5.3 ¾ thickness 515 442 7.6 A3 7010 T7651 48.92 ¼ thickness 514 451 5.7 ½ thickness 514 449 5.0 ¾ thickness 509 440 7.4 B1 7050 T7451 1.25 ¼ thickness 507 445 3.4 ½ thickness 519 470 4.6 ¾ thickness 507 428 5.6 B2 7050 T7451 8.83 ¼ thickness 513 446 4.2 ½ thickness 513 438 3.9 ¾ thickness 511 413 5.9 B3 7050 T7451 49.08 ¼ thickness 514 423 4.6 ½ thickness 505 420 4.8 ¾ thickness 513 442 3.7

K _IC toughness was also measured in LT and TL directions at ¼ thickness. The results, shown in Table 6, show that maturation has no significant influence on toughness. Table 6 Tenacity (in the final state) in the final state as a function of the aging time D for heavy plates in alloys 7010 and 7050 sheet metal Alloy / state Maturation time D [h] Location K _{IC (LT)} (MPa√m) K _{IC (TL)} (MPa√m) A1 7010 T7651 1.17 ¼ thickness 33.6 28.0 A2 7010 T7651 9 ¼ thickness 32.7 26.0 A3 7010 T7651 48.92 ¼ thickness 32.9 27.7 B1 7050 T7451 1.25 ¼ thickness 32.2 26.1 B3 7050 T7451 49.08 ¼ thickness 32.3 27.7

Example 2

Three AA7475 alloy rolling plates were processed by homogenization, hot rolling to a thickness of 46 mm, quenching and controlled pulling. The metallurgical state of the three products C1, C2 and C3 thus obtained was the state W51. For all three products, all manufacturing parameters were nominally identical and well controlled, and the only difference was the waiting time D between the end of quenching and the beginning of tensile stress relief.

Table 7 shows the stored elastic energy of the various sheets obtained, determined in the final state (ie after controlled pulling). When reducing the waiting time D between the end of the quenching and the beginning of tensile stressing, a reduction in the overall stress level W _L , W _LT and W is observed. Table 7 Elastic energy stored according to ripening time D for flat plate alloy 7475 W51 sheet metal Alloy / state Maturation time D [h] W [kJ / m ³ ] W _L [kJ / m ³ ] W _LT [kJ / m ³ ] C1 7475 W51 1.75 2.24 1.6 0.64 C2 7475 W51 22.5 4.51 3.61 0.9 C3 7475 W51 48 5.18 3.97 1.21

Example 3

Two AA7449 alloy rolling plates were converted by homogenization, hot rolling to a thickness between 16.5 and 21.5 mm, quenching and controlled pulling, followed by tempering. The metallurgical state of the two products D1 and D2 thus obtained was the T651 state. For these two products, all manufacturing parameters were nominally identical and well controlled, and the only difference was the waiting time D between the end of quenching and the beginning of tensile stress relief.

Table 8 shows the stored elastic energy of the various sheets obtained, determined in the final state (ie after controlled pulling). When reducing the waiting time D between the end of the quenching and the beginning of tensile stressing, a reduction in the overall stress level W _L , W _LT and W is observed. The small difference between the thicknesses of the two products does not lead as such to a significant difference between their stress levels. Table 8 Stored elastic energy (in the final state) as a function of the aging time D for alloy plates 7449 T651 sheet metal Alloy / state Thickness [mm] Maturation time D [h] W [kJ / m ³ ] W _L [kJ / m ³ ] W _LT [kJ / m ³ ] D1 7449 T651 16.5 10.5 6.3 5.56 0.74 D2 7449 T651 21.5 3 4.17 3.66 0.51

This result confirms that even for an alloy of the Al-Zn-Mg type with a high zinc content, such as 7449, the total elastic energy can be very significantly reduced by decreasing the maturation period D.

Example 4

By industrial processes which differed only in the waiting time, batches of control plates according to the invention were prepared. The stored energy was measured. Then, we developed a mathematical model that calculates this stored energy according to the critical parameters of the manufacturing process. The values of the stored energy measured for the sheets according to the invention have been used to validate this mathematical model. Then, this same mathematical model was applied to batches of Al-Zn-Mg type alloy sheets obtained by methods according to the state of the art. The figure 8 shows the values of the stored energy of the sheets according to the invention (with D ≤ 1 hour) (open points) ("Optimized") and according to the state of the art (with D ≥ 8 hours) (black squares).

It is found that for a thickness of between about 60 mm and about 100 mm, the stored energy is maximum. The method according to the invention leads, for a given thickness, firstly to a reduction in the overall level of residual stresses (that is to say of the stored energy W _totat ) of approximately 50%, and of on the other hand, a significant reduction in the statistical dispersion of this value. The effect of the invention on the overall level of residual stresses is particularly remarkable for thicknesses of between 40 and 150 mm and even more clearly for thicknesses of between 50 and 100 or even 80 mm.

Claims (14)

1. Method for producing Al-Zn-Cu-Mg type alloy thick plates having a thickness greater than 40 mm comprising between 4 and 12% zinc, less than 4% magnesium and less than 4% copper, minor elements ≤ 0.5% each, and the remainder aluminium, said method comprising hot rolling, solution heat-treatment, quenching, controlled stretching with permanent elongation greater than 0.5% and ageing,
   characterised in that the elapsed time D between the end of quenching and the start of controlled stretching is less than 1 hour.
2. Method according to claim 1, wherein the elapsed time D is less than 30 minutes.
3. Method according to claims 1 or 2, wherein said alloy is selected from the group consisting of the alloys AA7010, 7050, 7056, 7449, 7075, 7475, 7150, 7175.
4. Method according to any of claims 1 to 3, wherein the thickness of said plate is between 40 and 80 mm.
5. Method according to any of claims 1 to 3, wherein the thickness of said plate is between 40 and 150 mm.
6. Al-Zn-Cu-Mg type alloy thick plate with a thickness of at least 60 mm comprising between 4 and 12% zinc, less than 4% magnesium and less than 4% copper, minor elements ≤ 0.5% each, and the remainder aluminium, which is hot rolled, solution treated, quenched, stretched with a permanent elongation greater than 0.5%, aged,
   characterised in that its total elastic energy is less than or equal to $$\mathrm{W}\left[\mathrm{kJ}/{\mathrm{m}}^{\mathrm{3}}\right]=\mathrm{0.54}+\mathrm{0.013}\left({\mathrm{R}}_{\mathrm{p}\mathrm{0.2}\left(\mathrm{L}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)$$

   

   
   and in that its total elastic energy is less than 1.5 kJ/m³.
7. Plate according to claim 6, characterised in that its thickness is greater than 100 mm and its total elastic energy is less than 1.0 kJ/m³.
8. Inspection lot or heat treatment batch of Al-Zn-Cu-Mg type alloy thick plates with a nominal thickness of the plates between 40 and 100 mm comprising between 4 and 12% zinc, less than 4% magnesium and less than 4% copper, minor elements ≤ 0.5% each, the remainder aluminium, in a solution-treated, quenched, stretched and aged temper, characterised in that the total elastic energy W (expressed in kJ/m³) of the plates displays a standard deviation less than or equal to $$\mathrm{0.20}+\mathrm{0.0030}\left({\mathrm{P}}_{\mathrm{p}\mathrm{0.2}\left(\mathrm{L}\right)}\left[\mathrm{MPa}\right]-\mathrm{400}\right)$$

   

   around an average value.
9. Inspection lot or heat treatment batch of thick plates according to claim 8, characterised in that said average total elastic energy value is less than W [kJ/m³] = 0.54 + 0.013 (R_p0.2(L) [MPa]- 400).
10. Inspection lot or heat treatment batch of thick plates according to claim 9, characterised in that said average total elastic energy value is less than 3 kJ/m³.
11. Inspection lot or heat treatment batch according to claim 9 characterised in that said average total elastic energy value is less than 2 kJ/m³, and preferentially less than 1 kJ/m³.
12. Inspection lot or heat treatment batch according to any of claims 9 to 11, characterised in that the plates are made of alloy selected from the group consisting of AA7010, 7050, 7056, 7449, 7075, 7475, 7150, 7175.
13. Inspection lot or heat treatment batch according to any of claims 9 to 12, characterised that it consists of at least 3 plates and preferentially at least 5 plates.
14. Use of plates according to any of claims 6 to 7 or of a heat treatment batch of plates according to any of claims 8 to 14 for the production of machined components.

Images