WO2006136708A1

WO2006136708A1 - Method for continuously controlling the thickness of the base of extruded blanks of aerosol containers

Info

Publication number: WO2006136708A1
Application number: PCT/FR2006/001414
Authority: WO
Inventors: Hervé STOPPIGLIA
Original assignee: Cebal Aerosol France
Priority date: 2005-06-24
Filing date: 2006-06-22
Publication date: 2006-12-28
Also published as: DE602006002988D1; EP1899087A1; ATE409530T1; EP1899087B8; EP1899087B1; FR2887622A1

Abstract

The invention concerns a method for measuring the base (120) of impact-extruded cylindrical containers, based on the measurement of the temperature of the extruding equipment comprising a mobile part (10; 20) with a punch (10) actuated by a mechanical system (20) articulated from a reference point (23), said extruding equipment further comprising a passive part (30; 33, 34; 35) typically with a matrix (30) and support equipment (34; 35). The invention is characterized in that said temperature measurement is performed in a point of said passive part of the equipment, and in that, for a specific geometry of the container and a specific alloy, the thus measured temperature T is associated with the imposed production rate C and position X of said reference point (23) to estimate the thickness e obtained using a relationship R(e,T,X,C) = 0. Once established, said relationship enables the thickness base of the containers being manufactured to be controlled.

Description

PROCESS FOR CONTINUOUSLY CHECKING THE THICKNESS OF THE BOTTOM OF THE THREADS OF AEROSOL BOXES

TECHNICAL AREA

5

The invention relates to the manufacture of metal housings, in particular for aerosol dispensers, by shock-spinning. It relates more particularly to the continuous control of the thickness of the bottom of the shock-spun blanks which, after a possible stretching, are then conifted to produce the metal housing / o in one piece, typically aluminum alloy, distributors aerosols. This method can be applied to the online control of any part obtained by spinning.

STATE OF THE ART

/ 5

The blanks of the aerosol containers are in the form of boxes having a cylindrical wall open at one of their ends and connected at their other end to a bottom generally flat and perpendicular to the axis of the cylindrical wall. This blank is generally obtained by spinning of 0 cylindrical pins in an operation called "inverse spinning" by the blacksmiths but we prefer to call here "back extrusion", the term "inverse spinning" being used by the spinners to designate a different shaping operation, where the metal, in the form of a billet carried at high temperature and confined in the bore of a container, is urged to pass through

The central orifice (or several orifices located near the center) of a die which penetrates into said container to produce metal profiles of great length.

Impact spinning of aluminum alloy cylindrical pins is a process

30 different: admittedly, the material is also confined in a cavity of the tool, it is solicited by a moving part of the tool to flow to trαvers an opening on this part of the tool, but this opening is a peripheral opening, so that the radial component of the flow of the material is centrifugal and not centripetal. In addition, the operation is carried out without preheating the pieces, with a speed of

5 deformation significantly higher (the tool is operated by a crankshaft system - not crankshaft and using a hydraulic press). The operation is carried out at high rates on thick cylindrical pins which are brought one by one, at room temperature, to the spinning press. The latter, thanks to the crankshaft-crankshaft system, imposes rapid movements w alternating with a punch that suddenly hits each of these pieces, which are then evacuated as soon as they are in complete form.

Figure 1 shows the operating principle of a conventional "shock" spinning press. The spinning press has two parts: an active part

/ 5 comprising the punch 10 actuated by a rod system 20 and a fixed or passive portion 30, which comprises an anvil 31 (located directly under the punch) and a die 32 which acts as a fret to the anvil 31 - the set anvil and matrix will be called later under the generic term "matrix" 30 - and support tools, typically a grain 33 in support

20 of the anvil 31, a matrix holder 34 enclosing the grain 33 and a stack 35 of support wedges.

When the punch 10 is far from the die 30, the pin 1 is placed in the cavity 5 of the die. The spinning results from the sudden displacement of the

Punch 10 in the direction of the die. The latter hits the pawn and the spatial configuration offered by the punch and the matrix imposes on the metal, at the moment of impact, a large plastic deformation which results in the production, by a rear extrusion, of a hollow casing 100 with a wall 110 and a bottom 120. The diameter of the housing is generally close to

Initial diameter of the pion. The slenderness (height / diameter) of the case is generαl between 1 and 5. The bottom is generally flat and its thickness is thinner than that of the starting pawn.

The punch is animated by an alternating axial movement, its end passing from a "high point" close to the matrix to a "lower point" further away. The reciprocating movement is ensured by means of a rod 21 articulated on the one hand on the end 11 of the punch and on the other hand on a movable axis 22 also connected to a reference point 23 and to a crankshaft 26 rotating around a fixed axis A.

The movable axis 22 is not fixed directly to the crankshaft 26 but is connected thereto via a rigid rod 24 whose ends are articulated on said movable axis 22 and on a crankpin 25 of the crankshaft 26. It is also connected to a reference point 23 whose positioning is in principle fixed but adjustable at any time, for example by means of a wedge-screw system 40. The position of the "high point of the punch and therefore that of the reference point 23 have a strong influence on the thickness of the bottom of the spun product. It can be changed at any time, even when the crankshaft rotates.

This method is generally used to make the blanks of the aerosol dispenser housings. The bottom of these blanks remains generally flat because it is often at the level of the coniferous that it is shaped inverted dome. The shape of the bottom is designed so that the latter has on one hand a stable base and on the other hand resists the pressure difference between the outside and the inside of the aerosol dispenser. Thickness that is a little too low may cause the bottom to "turn over" and should be avoided.

It has been found that this thickness evolves during the in-line manufacture of the housings and this evolution was attributed in the first place to the progressive temperature rise of the tooling resulting from the heating of the pions which undergo a large and very rapid plastic deformation. As the thickness of the bottom of the blank evolves on the other hand very little during the - AT -

Final shaping (inverted dome), it is deduced that, to avoid the inversion of the bottoms of the aerosol cans, it is important to control the thickness of the bottom of the roughcast spun by shock, and it is necessary to make such a rapid control as possible to avoid, due to production rates, a

A large number of rejects each time a dimensional drift is noted. A solution consisting of removing the extruded blanks from the production line, or taking conified boxes, then cutting them and measuring for example by palmer the thickness of their bottom is too slow. It should be possible to perform a continuous measurement that would give as quickly as possible

IO an indication on a possible dimensional drift. No known means currently allows a direct measurement at 100% of the thickness of the funds of the spun products. It is therefore necessary to resort to indirect measurements, if possible not requiring any additional specific manipulation of the housings.

/ 5

A first solution has been tried. It relied on the correlation between the bottom thickness and the tooling temperature. It consisted of directly controlling the temperature of the punch. The expansion of the punch is indeed an important parameter affecting the final thickness of the bottom. Rapid 0 magnitude estimation shows that a simple rise of 30 ⁰ C of the average temperature along a punch of 300 mm corresponds to an elongation of ^{30 * 12 * 10- 6 * 300} = 9.6 10 " ² mm is almost 100 microns Unfortunately the thermocouples are fragile and, if mounted in the punch, they do not withstand the brutal mechanical punch.

Even before their break, the vibrations generated by the cycles of the press cause disturbances on the connection so that the results of the measurements are sometimes difficult to exploit.

The applicant has therefore sought another means for controlling the thickness of the housing bottoms, based on a parameter that is easy to measure continuously on the packages leaving the press or on the press itself. DESCRIPTION OF THE INVENTION

A first object according to the invention is a method for measuring the thickness of the bottom of the cylindrical casings spun by impact, based on the measurement of the temperature of the spinning tool, said spinning tool comprising a movable part with a punch and a passive part with a matrix, characterized in that said temperature measurement is performed at a point of the passive part of the tool, and in that, for a given production line, a given housing geometry and a given alloy, the temperature thus measured is associated with the imposed production rate and the position of a reference point associated with the moving part of the tool to estimate the thickness obtained. The punch performs reciprocating movements in a given direction using a mechanical system, typically a linkage system associated with a crankshaft, articulated from a reference point which is the link between said mechanical system and the frame of the spinning machine. The position of said reference point is defined with respect to said direction, which will be referred to hereinafter as "axial direction".

It is a method of indirectly measuring the thickness of the bottom of the spun products which involves a temperature measurement not on the punch but on the passive part of the tool, at a point distant from the zone impact of the punch, particularly at a point in the die, preferably at the anvil, or grain. This temperature-sensing zone is moved away from the shaping zone by a distance of the order of the diameter of the product to be spun. Preferably, a measurement point located in the grain is chosen near the interface with the anvil, typically at a distance of between 1 and 20 millimeters from said interface. However, this temperature measurement is not enough and it is associated with it, to obtain a good correlation with the background softener, the production rate, itself directly related to the speed of rotation of the crankshaft.

It is thus possible to establish, for products spun with a given alloy, having given dimensions (diameter, height) a relation of the type:

R (e, T, X, C) = 0 where e is the thickness of the bottom, T is the temperature measured at a given point of the tooling, X is the axial positioning of the reference point, defined with respect to a fixed point of the spinning press, that is to say, typically an anchor point of the machine, and C is the rate of production.

Contrary to all expectations, this relationship is valid regardless of the thermal regime of the device: at the start of the line, when the tool is completely cold, transient or established thermal regime. The heating due to the deformation of the pions results in the creation of a source which emits, at regular intervals, a heat flow which is evacuated in part by "convection" (the housings themselves which leave the spinning press ) and partly by conduction in the tooling. The established regime corresponds to the equilibrium between the heat flux created at a constant rate and the dissipated heat flow. After a certain number of cycles and at a certain distance from the shaping zone, it results in a stable temperature profile in the tooling. It is in this part of the tooling far from the shaping area that a continuous measurement of the temperature proves effective and provides reliable information. What is surprising, however, is that the information is also sufficient to operate an automatic thickness control system, even if the thermal regime is not yet fully established, in that it allows to obtain some control over the thickness faster than an operator can do, which during this transitional phase has insufficient information and can act only blindly. This good correlation is explained after the fact irrespective of the thermal regime of the system in that the dead time due to the conduction in the tooling of the heat flows generated by the self-heating of the pions is of the order of about fifteen seconds. The rise in temperature of the tooling 5 during this dead time is a few degrees Celsius and causes a decrease in the bottom thickness of at most a few tenths of a micrometer, which is very small in comparison with the differences noted in the conditions. industrial manufacturing on the thickness of the bottom that we want to control.

In practice, it is found that even in the case where the thermocouple is placed in the grain, that is to say far enough away from the box forming zone (distance of the order of magnitude of the diameter of the spinning pawn), this relationship proves effective. Despite the remoteness of the measuring point, a variation in the production rate results in an increase in

The immediate temperature at the punch is detected rapidly, that is to say after about ten seconds in the grain.

To establish this relation R (e, T, X, C) = 0, for example, an experiment plane is engaged on the production line, in which the

20 parameters: C production rate, X position of the reference point. The thickness e, obtained in steady state, is measured by palmer on the extruded blanks. The temperature T is measured at the chosen point, located in the passive part of the tooling, preferably at a distance corresponding approximately to the diameter of the spun blank, typically in the

25 at a point in the vicinity of the anvil.

The most appropriate relationship, established in the experimental design and during the process development tests, resulted in an equation giving the bottom thickness as the sum of four independent terms: 30 e = ao + f (T) + g (X) + h (C) where αo is a constant for resetting after dismantling a tool and then mounting a new tool and where f, g and h are monotonic functions of a single variable.

The bottom thickness is thus reliably connected by a sum of non-interdependent monotonic functions of distinct unique variables: f is a decreasing monotonic function of the temperature T measured at a point in the passive part of the tooling, g is a monotonic function of the position of the reference point of the linkage, decreasing if the longitudinal axis used to define the X coordinate is oriented from the reference point to the matrix, h is a monotonic decreasing function of the production rate. The functions f, g and h are defined once and for all for a given production line, a given spun gear geometry and an alloy given, for example, using the experimental plane mentioned above.

The coefficient ao is an adjustment parameter estimated at each tool replacement. It is necessary to wait for the press to operate in steady state, that is to say typically 10 to 15 minutes of walking in constant cadence after the change of tools. The spun products are then taken and the thickness of their bottom is measured directly, for example by cutting the bottom and making a measurement by palmer.

Once established, this relation makes it possible to define a reference position Xc of the reference point corresponding to the target thickness ei to obtain with a given rate Ci and a given temperature Ti. The reference position Xc of the reference point is defined by:

Xc≈ g-Mei -l ao + fCTiJ + hlCi))) where g- ¹ is the inverse function of g.

Another object of the invention is a method of controlling the thickness of the bottom of cylindrical casings spun by shock, based on the measurement of the temperature of the spinning tool, said spinning tool comprising a moving part with a punch moved by a mechanical system articulated from a reference point and a passive part with a matrix, characterized in that: a) we use the process for measuring the bottom thickness of the impact-spun cylindrical casings described above and for a given production line, a given casing geometry and a given alloy, a relation R (e, T, X) is established , C) = 0 where e is the bottom thickness, T is the tooling temperature, X is the reference point position and C is the production rate; b) a measurement is made of the tooling temperature T at the measurement point located in the passive part of the tool and, taking into account the production rate C at the time of the measurement, the value of the reference position Xc of the reference point corresponding to a target thickness ei, Xc being deduced from the relation R (ei, LXcC) = O; c) measuring the position X of said reference point; d) comparing said position X of the reference point with said setpoint position Xc and, to obtain on the following pins a thickness closer to the target thickness, defining a corrective action depending on the positional deviation noted in FIG. following this comparison.

The corrective action is based on the position deviation observed between the reference position Xc of the reference point deduced from the temperature measurement and the value of the manufacturing rate at the time of taking the temperature - its value is given for example, if we use the monotonic functions f, g and h presented above, by Xc = g - '(ei - (ao + f (T) + h (C))) - and the real axial position X, measured, from the reference point. The corrective action can be either a correction on the positioning of the reference point or a correction on the rate of production. It goes without saying that the production rate should, as far as possible, be as high as possible and therefore not an ideal parameter for regulation. Of Therefore, preference will be given to the positioning of the reference point by regulating with respect to the set point value Xc of the reference point indicated above.

According to a preferred embodiment of the invention, the corrective action consists in imposing on the reference point a displacement a (X-Xc) in the axial direction, a being in absolute value a coefficient close to 1, typically between 0.5 and 1, which depends on the correction dynamics of the motor used for the system controlling the axial displacement of the reference point, for example the wedge-screw system described in the example below.

Referring to the diagram of FIG. 1, it can be seen that the correction on the position of the reference point is preferably done by acting on a wedge / screw-type system such as that referenced 40 or any other equivalent system making it possible to control or to enslave an axial displacement. The assembly is fixed on a fixed part of the frame of the machine by means of a base 42. A screw 43 is rotated, for example with the aid of a servo motor, and the screw moves a wedge 44 in a direction perpendicular to the axis of the punch. By moving, this wedge acts via its oblique face on the position of the part 41 on which is fixed the reference point 23.

To act effectively on the position of the reference point, it is imperative to eliminate the uncertainty of the clearance between the thread of the screw 43 and the complementary thread of the nut formed in the corner 44. For this reason, we do not It does not merely control the angular position of the screw 43, but a displacement sensor (not shown in FIG. 1) is provided between the base 42 and the part 41, which provides an accurate indication of the position of the reference point 23.

Industrial line tests have shown that, in practice, a single relationship is sufficient for the same line, irrespective of the alloy used and that it was valid for several geometries (diameter, height) of the spun product. It is thus possible to control the manufacture of housings whose bottom has a thickness which varies within a range of values less than 15/100 th of a millimeter.

5

Particular embodiment of the invention (FIG. 3)

Continuous control of the bottom thickness of 53 * 204 1050 alloy aerosol can blanks. W

The shock-spun blanks are obtained from 1050 alloy pegs (reference according to the Aluminum Association standard) with a thickness of 6.5 mm.

The temperature measurement is carried out using a thermocouple 50 in a point 36 of the passive part of the tool, away from the impact zone of the punch, and located at the level of the grain 33. More precisely it is 10 mm below the interface between the anvil 31 and the grain 33.

An experiment plan was made on the production line, varying the parameters C rate of manufacture and X position of the reference point 23. For each blank of housing retained to perform a measurement of thickness, it was noted temperature T at the measuring point 36, the production rate C and the axial positioning X of the reference point at the moment when it was spun by shock. This led to a relationship of the type:

E = ao + f (T) + g (X) + h (C)

Typically, for a given production line, for a given box geometry and for a given alloy, the experiment plan makes it possible, with a

30 readings (each having been measured bottom thickness of at least four extrusions spun), specify the form of the functions f, g and h figurαnt in the equation e = ao + f (T) + g (X) + h (C). Measurements were made as early in the manufacturing process, when the tooling is still cold after several tens of minutes, when the tool has reached, at a given rate, a stable thermal profile.

The thickness is regulated by acting on the position of the reference point 23. The principle consists in calculating a reference position Xc of the reference point by using the model developed for estimating the thickness. The equation giving this setpoint position is easily obtained:

X _c = g- '(ei - ( _ao + f (T) + h (C))) the functions f, g and h having been defined thanks to the experimental design carried out once and for all on the production line given, the coefficient of registration ao being defined at each renewal of tooling and the thickness ei being the thickness referred to.

During manufacture, the temperature is continuously measured at point 36 of the passive part of the tooling using thermocouple 50.

X from the reference point 23 by means of a linear displacement sensor and this value is compared with the reference value Xc. The deviation X-Xc is used as a correction parameter to act on the position of the reference point via the corner-screw system 40. The precise positioning of the reference point 23 is controlled by means of a sensor Linear displacement type LVDT which connects the base 42 to the part 41. This slaving has the advantage of not requiring an in-depth transformation of the control automaton of the production line which already has a system for measuring rates and positioning. It is enough to add a measurement of temperature, to improve the control of the positioning by means of a sensor of linear displacement judiciously placed and to introduce a software of regulation based on the model of estimate of the thickness exposed above.

Claims

1) Method for measuring the thickness of the bottom (120) of cylindrical casings spun by impact, based on the measurement of the temperature of the tooling of

Spinning, said spinning tool comprising a movable part (10; 20) with a punch (10) performing reciprocating movements in an axial direction (200) by means of a mechanical system, typically a linkage (20) associated with a crankshaft ( 26), articulated from a reference point (23), said spinning tool also comprising a passive portion w (30; 33,34; 35) with typically a die (30) and support tools (34). 35), characterized in that said temperature measurement is performed at a point (36) of said passive part of the tool, and in that, for a given production line, a given box geometry and a given alloy , we associate the temperature T thus measured with the rate

C and the axial position X of said reference point (23) for estimating the thickness e obtained by means of a relation R (e, T, X, C) = 0.

2) Measuring method according to claim 1, characterized in that said temperature measurement is performed at a point typically located at a

Distance of the order of the diameter of the product spun and belonging either to the matrix (30) (anvil (31) or peripheral portion (32) enclosing said anvil), or to a support tool (33; 34; 35) , for example the grain (33) located under the anvil (31).

3) A measuring method according to claim 1 or 2, characterized in that said temperature measurement is performed in point (36) of the grain (33), located in proximity, typically between 1 and 20 millimeters, of the interface with the anvil (31).

4) Measuring method according to claim 1, characterized in that said relation R associating the measured temperature T, the production rate imposed C, the position X of the reference point (23) and the thickness e of the bottom of the spun product is a relation of the form: e = ao + f (T) + g (X) + h (C) where ao is a constant used for resetting after mounting a new tool and where f, g and h are monotonic functions of a single variable, established once and for all, for example by means of an experimental design, for a given production line, a given spun product geometry and a given alloy.

Measuring method according to claim 4, characterized in that said relationship associating the measured temperature T, the imposed production rate C, the position X of the reference point (23) and the thickness e of the bottom of the spun product is a relation of the form: e = ao + f (T) + g (X) + h (C) where ao is a constant for resetting after mounting a new tool and where f, g and h are monotonic functions of a single variable, established once and for all for a given manufacturing line, regardless of the alloy, diameter and height of the spun cylindrical case.

A method of controlling the thickness of the bottom of the spun cylindrical casings, based on the measurement of the temperature of the spinning tool, said spinning tool comprising a moving part with a punch driven by a mechanical system (20). and 26) articulated from a reference point (23) and a passive part with a matrix, characterized in that: a) the method of measuring the bottom thickness of the impact-impacted cylindrical housings according to any one of claims 1 to 5 and for a given production line, a given housing geometry and a given alloy is determined, a relation R (e, T, X, C) = 0 where e is the thickness of the background, T the tooling temperature, X the position of the reference point (23) and C the rate of production; b) a measurement is made of the tooling temperature T at the measuring point (36) located in the passive part of the tooling and, taking into account the production rate C at the time of the measurement, the value of the reference position Xc of the reference point (23)

5 corresponding to a thickness referred to ei, Xc being deduced from the relation

c) measuring the position X of the reference point (23); d) comparing said position X of the reference point (23) with said setpoint position Xc and, to obtain on the following pieces a thickness

/ o Closer to the target thickness, a corrective action is defined depending on the position deviation observed as a result of this comparison.

A method of controlling the bottom thickness of the impacted cylindrical casings according to claim 6 wherein said corrective action is to move the reference point (23) in the axial direction (200).

8) A method of controlling the bottom thickness of the impacted cylindrical casings according to claim 6 or 7 wherein

A) the relation R (e, T, X, C) = 0 associating the measured temperature T, the imposed production rate C, the axial position X of the reference point (23) of the linkage (20) and the the thickness e of the bottom of the spun product is a relation of the form: e = ao + f (T) + g (X) + h (C)

Where ao is a constant for resetting after mounting new tooling and where f, g and h are monotonic functions of a single variable; b) the reference position Xc of the reference point (23) deduced from the measurement of temperature T of the tooling at the measurement point (23) and of the production rate C at the time of taking the temperature is

30 given by the relation:

Xc = g- ^] [e _] - [oo + f [J) + h (C))) where g- 'is the inverse function of gc) the deviation X-Xc is used as a correction parameter to act on the reference point position through a system controlling the axial displacement of the reference point (23). ), Typically a wedge-screw system (40) with screw (43) servo-controlled.

9) A method of controlling the bottom thickness of the impact-spun cylindrical casings according to claim 7 or 8, wherein the precise positioning of the reference point (23) is controlled by means of a displacement sensor.

/ o Linear connecting the base (42) to the piece (41).

A method of controlling the bottom thickness of the impacted cylindrical packages according to any one of claims 7 to 9 wherein said corrective action is to impose on the reference point (23) of

The linkage (20) a displacement a (X-Xc) in the axial direction (200), a being a coefficient whose absolute value is close to 1, typically between 0.5 and 1.