WO2007093481A1

WO2007093481A1 - A process for manufacturing ceramic tiles

Info

Publication number: WO2007093481A1
Application number: PCT/EP2007/050558
Authority: WO
Inventors: Pietro Rivola; Alessandro Cocquio; Enrico Primo Tomasini; Gian Marco Revel; Paolo Pietroni; Giuseppe Pandarese
Original assignee: Sacmi Cooperativa Meccanici Imola Societa' Cooperativa
Priority date: 2006-02-13
Filing date: 2007-01-19
Publication date: 2007-08-23
Also published as: RU2008130659A; BRPI0707435A2; ITRE20060018A1; CN101378885A; EP1984156A1

Abstract

A process for manufacturing ceramic tiles by ceramic powder pressing in cavities of a die provided with an isostatic punch a membrane of which rests on oil contained in adjacent and non-communicating chambers in continuous operation, characterised in that it comprises following operations: detection of a density of the powders pressed in a plurality of points n uniformly distributed in each area of a tile which plurality of points corresponds to a chamber of of the punch; performing a calculation of a mean of density readings taken at all the points in each area; comparing read values with a desired optimal density value; removing or, respectively, adding oil to each isostatic chamber of the punch if the mean of densities in the chamber is above or respectively below the optimal density value.

Description

A PROCESS FOR MANUFACTURING CERAMIC TILES

Technical Field

The invention relates to processes for manufacturing ceramic tiles by powder pressing and subsequent firing at high temperatures.

Background Art

The powders known as ceramic powders destined to be pressed in the cavities of the dies have a moistness of about 6%, generally comprised between 4% and 8%.

These powders are supplied to the cavities of the die by means of a carriage provided with a bottomless frame, loaded with powder, which is positioned above the cavity of the die; the lower punch of the die is raised to the level of the upper edge of the cavity.

The subsequent descent of the punch by a quantity equal to the thickness of the soft powder to be pressed, causes a descent of the desired quantity of powder into the cavity, which powder is levelled by the carriage itself as it distances in order to enable the upper punch to descend. The die cavity comprises a lower punch which is solidly constrained to the die mechanisms, and an upper punch which is solidly constrained to the crossbar of the press.

When these punches are made of steel, irregular distribution of the powder in the cavity gives rise to different densities in the body of the tile, generating problems during the following firing stage, due to different shrinkage corresponding to different densities.

A remedy has been sought to this problem with the introduction of isostatic dies, in which at least one of the punches is constituted by an elastic membrane resting on a liquid which fills a closed chamber of the membrane. Elimination of the irregularities in the density leads however to shape defects, since the thickness of the tile corresponding to the zones exhibiting a defect in the powder density is smaller than the nominal thickness, and vice versa. Shape defects have been corrected using solutions which are extraneous to the present invention and which are therefore not cited herein. However, there is a further problem originating in the fact that the density is responsible for the shrinkage of the tile in its entirety, resulting in the fact that the fired tile is smaller than the unfired tile.

This fact of shrinkage is obviously taken into consideration during the die design stage, so that the fired tile will be of the desired dimensions. The prior art describes dies having several cavities, each of which cavities comprises an isostatic punch. In these dies, the isostatic punch chambers are in communication in order to provide uniformity of density not only for the single tiles, but also for tiles made simultaneously in the die.

The above prior art, well known to experts in the field, has however been seen to be unsatisfactory due to the increased sizes of the tiles and the consequent reduction in their thickness in comparison to the surfaces thereof.

The density variations tolerated for the single tile must be contained within limits which are not possible with isostatic dies, due to the relevance assumed by other parameters. The density of the tiles depends also on the moistness level in the powder, which can vary, even if very slowly, within the time taken for a continuous production of a batch of tiles, with the consequence that even if the density within a same tile remains constant thanks to the isostatic pads, the tiles from the same batch might during firing be subject to different levels of shrinkage due to changes in the moistness of the powders.

It is known that oscillations in moistness values in powders used for realising ceramic products, given a same pressing thrust, will create variations in the density of the pressed product. These density variations lead in turn to shrinkage variations during the following firing stage. It has been proposed to vary the pressing force in accordance with the moistness in the powder at inlet, but this system has not shown itself to be sufficient to solve the problem because the density, given equal pressing force, is not affected only by the moistness, which acts in synergy with the distribution and granulometry of the powder, parameters that can change over time in the various areas of the die cavity, in spite of the use of isostatic dies.

The above considerations, in relation to a die having a single cavity, are further complicated for dies having more than one cavity.

Disclosure of Invention

The aim of the present invention is to provide a manufacturing process for tiles which guarantees keeping shrinkage to a very low level during firing of a pressed unfired tile, over a time-span of the manufacture of a whole batch of tiles.

Since shrinkage during firing, with an equal firing temperature, is substantially conditioned by the density of the pressed powder, in order to keep the shrinkage within the desired limits the invention proposes a control process of the pressing operation which process takes account of the actual thicknesses and density values, detected at various points of each single tile downstream of the press. The object of the present invention is a control process which allows continuous control of the parameters which influence shrinkage, and which continuously corrects these parameters.

As the parameters vary very slowly over time, the process of the invention controls, in the tiles exiting the kiln, the local density in a certain number of points therof and uses the detected density values, or other values which in the density measuring system are biuniquely associable to the density, in order retroactively to control the operating parameters of the press and/or the die.

In consideration of the slow shift that the above-indicated parameters undergo in an industrial plant working in continuous cycle, the invention adopts, as sensitive elements for the correction of the parameters of the -A-

press/die not the single detected values, but the relative mean values, which change over time.

In the case of dies having more than one cavity, the invention further provides that the membranes of the isostatic punches of each cavity rest on chambers which are hydraulically non-communicating with the punch chambers of the other cavities.

The same is true in the case of dies having a single large-size cavity which use an isostatic punch the membrane of which rests on a plurality of adjacent chambers, which chambers must not be in communication. The density detector is thus able to provide a mapping of the local density values of the tiles as they exit the press.

The values detected, or rather the shifting means thereof, are used retroactively to modify the functioning parameters of each single cavity of the die, or of the single non-communicating chambers of the punch of a large- size die, as will better emerge from the description herein below.

This allows trends in the variations to be detected and at the same time allows the detected values to be cleaned of any incorrect values caused accidentally.

The subordinate problem of maintaining the density constant over time, for example during the manufacture of the whole batch of tiles, in spite of the slow variation in the moistness of the tiles, is solved by the invention which at predetermined intervals measures the moistness in the powder supplied to the cavities of the press, and uses the detected values in order to change the functioning parameters of the press, such as for example the pressing thrust. The further subordinate problem of an irregular distribution of the powders in spite of the action of the isostatic punch is resolved by the invention which uses a density measuring system that also uses the measurement of the thickness of the tiles in the points subjected to the detection process. To this end the invention describes the use of non-contacting ultrasound systems for detecting the density; these systems, as known, include the use of the ultrasound wave length velocity in the material of the tile, and thus detect the time required by the ultrasound wave in order to cross the thickness of the tile, which time, all other parameter values being equal, is certainly proportional to the local thickness of the tile at the point crossed by the ultrasound wave. With the tile thickness already known, the time measurement enables the velocity of the wave travelling through the material to be derived; this can be correlated to the density of the tile material.

This correlation is derived during calibration of the system in relation to the moistness. For example, a graph is generated containing the curves representing the density according to the velocity for each moistness value. It has been found that there exists a ratio of linear proportionality expressed, for each moistness value, by a known and constant inclined line in the density/velocity diagram. If one of these lines is known, the other lines relating to different moistness values are part of a bundle of parallel lines each identifiable by a single point. In order to measure the thickness of the tile at the point of investigation into the density thereof, in a case where local thickness variations are presumed which are such as to influence the density values significantly, the invention includes measuring the thickness using a no-contact system, such as a triangulating laser system.

The detected values of dimensions susceptible to biunique correlation procedures are used to retroactively correct the operating parameters of the means supplying the powder to the die cavity, and precisely the descent of the upper punch and/or the velocity of the powder supply carriage.

In this case too it is advantageous to use the shifting means of the detected values instead of the single values thereof. With the shifting density and thickness means available for various points of the tiles realised in the various die cavities, various situations can be identified, and for each different situation the correct action can be taken retroactively on the parameters concerned. An example of actuation of the process of the present invention is described herein below.

Best Mode for Carrying Out the Invention.

EXAMPLE The process is applied to a press having several outlets, i.e. equipped with one die having Z cavities each provided with isostatic punches. In the present example Z is equal to 3.

From the above-described press, three lines of tiles exit which are parallel to one another and arranged each on a conveyor line, which for ease of description we will denote using the letters of the alphabet A... Z, in this case A, B and C.

Before starting continuous production of a batch of tiles the operator must manually regulate the press and the die, in order to optimise the thicknesses within a single tile and also among the tiles to be pressed in the other cavities A... Z.

The operator therefore defines the range of admissible values, i.e. the range within which the measured values must fall in order to guarantee the quality of the tiles. This range of admissible values is sent to the press control processor and to the means for supplying the powder, together with the admissible powder moistness value range, generally comprised between 4% and 8%, and the optimal density value for the pressing, expressed as T_Oττ- T_OTT is statistically the same as the mean T_M of the values T_x detected in a certain number of tiles. At this point continuous production is started up, together with the data detection processes.

In the powder supply hopper, at a point close to the outlet conduit thereof, a usual-type moistness measurement device is installed, which provides the percentage values of a parameter U% biuniquely linked to the powder moistness. These percentage values are sent on to the control processor. An ultrasound device is installed on each line A... Z which device can monitor the density in a plurality of points in a same tile; possibly a laser device is also present, for non-contact measurement of the tile thickness at the density investigation points. The ultrasound device operates according to known principles, and provides an indication of the density derived from the time needed by the ultrasound wave to cross the thickness of the tile.

One of the useful parameters T_x can be the tile crossing time, which we shall denote by T_x. The value T_x is directly proportional to the density. The device can be activated as each tile passes, or can advantageously be activated at regular and predetermined intervals.

For each tile T_x is taken at n points, in the example nine points regularly distributed on three lines which are parallel to the front and back sides of the tile (by "front" and "back" meaning in the advancement direction). The points close to the front side shall be points 1 ...y, in the example points

1 , 2, and 3, while the points close to the back side shall be, in the example, points 7, 8 and 9.

From the tiles of line A, nine values will be taken: T_A, i T_A,g, which are sent on to be stored in the memory of the control processor. The same exercise is performed on all the lines A... Z.

In the following explanation the small characters will refer to three lines, A, B and C, and to nine detection points; extension of these in the case of Z lines and n detection points is readily done.

At regular invervals the control processor performs the following operations: - it calculates the mean value for each of the nine values T_A,i ....T_A,9 of the last five values received, obtaining nine values for each line, which we shall

- it calculates the mean of values T_M,A,I - - - T_M,A,9 obtaining the value T_A,M for line A; the same is done for lines B and C. - it calculates the mean of values T_M,A,I - - - TM, A, 3, obtaining the value which we shall call T_M,A,I3- - it repeats the operations for lines B and C.

- it calculates the mean of values T_M,A, i- -- T_M,A,9, obtaining the value which we shall call T_M,A,79-

- it calculates the mean of values T_A,M, T_B,M and T_C,M, obtaining the value which we shall call T_M.

- it calculates the mean of values T_MAI3- - - T_Mci3 and values T_MA79-- T_MC79, obtaining the values T_M, ant and T_M, _post-

The control processor's intervention on the press occurs as follows.

A. If one of the values T_A,M, T_B,M and T_C,M is greater or, respectively, lower beyond a predetermined amount than T_M, the processor commands a device, of known type, connected independently to the isostatic punch of each cavity, which removes or respectively injects a measured quantity of oil from or into the respective punch.

The operation is repeated up until the values T_A,M , T_B,M and T_C,M differ from T_M by a lower entity than the predetermined entity.

B. Once a homogenization of the density values of lines A, B and C has been obtained, the processor compares the mean density T_M,ant of the front portions of the tiles with the mean density T_M,_Post of the back portions thereof. If T_M,ant is greater, or respectively lower, than T_M,_Post, the processor commands the special means for vertically moving the lower punches of the die in order to start the descent when the scraper of the loading carriage is in a more retreated position, or respectively in a more advanced position with respect to a front rest position thereof. This operation is repeated up until when the difference between T_M,_ant and T_M,_Postdoes not fall within a desired interval.

If after a predetermined number of operations this does not happen, the processor reduces, or respectively increases, the advancement velocity of the frame of the carriage into the space close to the front rest position in the shuttling motion thereof during the outward run thereof. This results in a homogenization of the tile density in the motion direction of evacuation from the die, with the mean value of density of the front portions TM,ant differing from the mean value of density of the back portions T_M,_Post, in the present example T_MA79 by an entity which is lower than the maximum admissible error.

C. The process ascertains that the value U% of the moistness in the powder is comprised within a predetermined interval based around the initial value; if this is so, the processor compares the mean value of the densities detected in the tiles with the value T_Oττ considered as optimal.

Should the mean value T_M be above or respectively below the value T_Oττ, the thrust of the press is reduced or respectively increased up until the desired value is reached.

It is stressed that the corrections made at each intervention are of small entity, the retroactive process being repetitively activated until the detected values differ from the desired values by an equally small amount.

The above-described process means that after firing all the tiles differ from each other by a lower quantity than required in order to be considered all of the same calibre.

Claims

Claims.

1 ). A process for manufacturing ceramic tiles by ceramic powder pressing in cavities of a die provided with an isostatic punch, a membrane of which punch rests on oil contained in adjacent and non-communicating chambers, in a continuous operation, characterised in that it comprises following operations: detection of a density of the pressed powders in a plurality of points n uniformly distributed in each area of a tile, which plurality of points corresponds to a chamber of the punch; performing a calculation of a mean of density readings taken at all the points in each area; comparing detected values with a desired optimal density value; removing or, respectively, adding oil to each isostatic chamber of the punch if the mean of densities in the chamber is above or respectively below the optimal density value. 2). The process of claim 1 , characterised in that the density value is expressed by means of a value which is biuniquely correlated to the density as a function of the moistness.

3). The process of claim 2, characterised in that the said value is the velocity of an ultrasound wave travelling in the pressed powder of the tile. 4). The process of claim 3, characterised in that the density is measured by an ultrasound device and is expressed by a value which is correlated to the velocity of the ultrasound wave, such as a time required by the ultrasound wave to pass through a thickness of the tile. 5). The process of claim 1 , characterised in that the value expressing the density is sent continuously to a processor which controls the device retroactively and which removes oil from or adds oil to isostatic chambers of a punch where an excess or lack of density has been revealed. 6). The process of claim 1 , characterised in that the isostatic die comprises at least two cavities, isostatic pads of which are provided with a single chamber and which are hydraulically isolated from one another. 7). The process of claim 6, characterised in that it comprises following further operations:

- it calculates the mean value for each of the n values _TA,i - ---T_A,_n detected in each cavity for the last m readings, obtaining for each cavity A... Z of die n values T_M,A,I - - - T_M,A,_Π and respectively T_M,B,I - - - T_M,B,Π, ■ ■■, T_M,Z,I T_M,Z,_Π ;

- it calculates the mean of values T_M,A,I - - - T_M,A,_Π obtaining the value T_A,M for line A, and in the same way calculates values T_B,M, T_C,M, T_Z,M for cavities B,

C....Z.

- it calculates the mean of values T_A,M- - - T_Z,M obtaining the value T_M assumed as indicating the optimal density value;

- it compares values T_A,M- - - T_z,M with value T_M , and - removes or respectively adds oil to each of the isostatic chambers of the punch of lines A, ...Z if values T_A,M- - - T_Z,M of the density are above or respectively below the mean value T_M ;

- it repeats the operation for a number of interventions which are sufficient to reduce the difference between T_A,M... T_Z,_M and T_M within predetermined limits.

8). The process of claim 7, characterised in that the operation of removing or adding oil is performed by a device which is retroactively commanded by the process to which the data T_x is sent for processing.

9). The process of claim 6, characterised in that it comprises following further operations:

- it calculates the mean value for each of the values T_M,A,i - ---TM,A,y ^{of tne} Y points of each tile close to a front edge thereof, deriving a value T_M,A,i_y and repeating the operation for all the lines A....Z;

- it calculates the mean value T_M,ant of the values T_M,A,i_y T_M,_z,-ι_y; - it calculates the mean of values T_M,A(n-_y+i)- -- T_M,A,_Π relating to (n-y+1 ) points of each tile close to the back edge therof, obtaining a value termed T_M,A(n-_y+i),n and repeating the operation for all the lines A... Z.

- it calculates the mean T_M, _post of values T_M,_A(n-_y+i)- -- T_M,z(n-_y+i), n, - it compares the mean density T_M,ant of the front portions of the tiles with the mean density T_M,_Post of the back portions thereof, and if T_M,ant is greater than, or respectively lower than T_M,_Post it advances a descend command or respectively retards a descend command to the means for lowering the lower punches of the die; - it repeats the operation for a number of interventions sufficient to reduce the difference between T_M,ant and T_M,_Post to within predetermined limits. 10). The process of claim 9, characterised in that the action to descend to the punches is commanded by the processor in retroaction, following the processing of the values Tx supplied by the density measuring device. 11 ). The process of claim 9, characterised in that it comprises a further operation of reducing, or respectively increasing, an advancement speed of the frame of the carriage, if following a predetermined number of interventions the difference between T_M,ant and T_M,_post does not fall within predetermined limits. 12). The process of claim 1 1 , characterised in that the further operation of reducing or respectively increasing the advancement velocity of the frame is controlled retroactively by the processor following processing of the values Tx supplied by the density measuring device. 13). The process of claim 1 , characterised in that it comprises a further operation of measuring the thickness of the tile in the points of density detection, for a more precise calculation of the speed of the ultrasound waves in the material.

14). The process of claim 13, characterised in that the thickness measuring operation is performed by a no-contact measuring device. 15). The process of claim 14, characterised in that the no-contact measuring device is different to an ultrasound measuring device. 16). The process of claim 14, characterised in that the no-contact measuring device is a laser thangulation apparatus.

17). The process of claim 1 , characterised in that it comprises a further operation of determining a correlation between a tile moistness and powder density, or a value indicative of the density.

18). The process of claim 17, characterised in that the value indicative of the density is a velocity of propagation of the ultrasound wave in the material.

19). The process of claim 18, characterised in that the correlation between the velocity of the ultrasound wave and the density of the pressed tile is experimentally deduced using a moistness value, and the representative curve thereof is approximated with a straight line characterised by an angular coefficient therefor.

20). The process of claim 19, characterised in that thereafter readings for velocity/density are taken for a congruous number of moistness values contained in a determined range of work, and straight lines are drawn through the points of the density/velocity plane using the angular coefficient found.