DK2394760T3 - Method and apparatus for regulating the metal dosage in die-casting cells - Google Patents

Description

The present invention relates to a method and a device for regulating the metering of metal when manufacturing a plurality of castings by means of a die casting cell according to the preamble of the main claim.

Die casting cells are used in the manufacture of castings by introducing a specific quantity of molten metal, called the metering weight, into a fill chamber and pressing this metal in a casting mould by means of a plunger, wherein the metal solidifies in the casting mould. To this end, die casting cells comprise a metal inlet in the form of a metering furnace or scoop system, a die casting machine that manufactures the casting to be manufactured by pressing it into a casting mould, a spray device for cleaning the casting mould, and a removal device for the casting, for instance, a robot. High metering accuracy is desirable in order to manufacture castings with a defined weight. Metering accuracy is defined as the ability of a die casting cell, especially the device for supplying metal contained therein, to re-producibly keep constant the mass of the metal introduced such that said mass does not exceed or fall below the limits of an established tolerance range. The smaller this tolerance range can be selected and the lower the number of castings that are outside of the tolerance, which are defined as waste and are rejected, the better the metering accuracy is.

Fluctuations in the metering rate, and thus in metering accuracy, may be divided into two categories: statistical metering fluctuations that have a Gaussian distribution, and systematic metering fluctuations that are expressed by to the metering weight drifting in certain directions or that are caused by external interventions such as cleaning or refilling processes .

The most accurate method for determining metering accuracy is to weigh manufactured castings, but this would mean an intervention into the production operation and would therefore increase production times.

Various other methods for regulating the metering weight and compensating fluctuations in metering are known from the prior art. DE 40 29 386 Ά1 discloses an integral method in which a pressure-over-time integral is used for determining the metering quantity. The furnace internal pressure is regulated via a single-stage regulator or via a cascade regulator, described for instance in DE 42 04 060 C2, as long as the metering process continues. In this case an indirect determination of the metering quantity using supply pressure and time is employed, but it suffers from the drawback that the compressed air trapped in the furnace is compressible and thus the metering quantity depends on different factors that generally cannot be measured and that can change during the course of operation. These include changes in metal deposits on the lip of a metering pipe, changes in the supply pressure as the furnace becomes empty, and hydrostatic increases in the pressure of the melt in the furnace due to filling during the metering process. For compensating the metering weight, the user of the metering furnace may manually alter metering weight, supply pressure, and supply time.

Another option for compensating the metering quantity is comprised in evaluating parts that are out of tolerance and revising the metering weight. The set me- tering quantity is changed if metering weights outside of the permissible tolerances have been found. However, this method does not lead to the desired success in minimizing the number of rejected parts because it takes into account individual events, but does not react to them until after they have occurred. EP 0 693 983 A1 discloses a smelting furnace in which a remaining press length is determined and a volume of molten material is set for another pressing process depending on the determined value of the remaining press length. A similar device is also known from EP 0 581 949 A1. A pressing device in which a remainder thickness is determined by a magnetic sensor is disclosed in JP 4017972 A. A corresponding measurement of a remaining press length with subsequent adaptation of a quantity of molten material for further pressing processes is known from JP 1027759 A. In a device disclosed in JP 4091855 A, a position of an end side of the remainder is detected and the remaining press length is also determined. Then an automatic correction of the quantity of the molten metal to be added to the press chamber is determined. JP 4 094 855 A discloses a device in which a remaining press length is determined and compared to a predetermined range. If the value determined for the remaining press length is outside the predetermined range, the injection is adapted for further pressing processes . JP 56131061 A discloses a device in which a remaining press length is calculated. If the remaining press length is outside of a certain range, a mean value is calculated, and if this mean value is also outside of the predetermined range, a correction is made to the quantity to be added.

In the device illustrated in JP 2003-112247 A, an adaptation of the metal to be added is undertaken if a weighed quantity of the metal exceeds a certain value .

The underlying object of the invention is therefore to develop a method and a device that prevents the aforesaid drawbacks, that is, with which it is possible to achieve greater metering accuracy automatically.

This object is inventively attained using a method for regulating metal metering having the features of the main claim and using a device according to claim 10 that is suitable for performing this method.

Advantageous refinements of the present invention are possible using the features provided in the dependent claims .

The invention provides a use of a so-called remainder as a measure for the metering quantity. A sprue on the manufactured casting that occurs where the metal is introduced into the casting mould and that after solidifying initially remains on the casting and is typically removed later in the procedure is called a remainder. The diameter of the remainder is defined by the plunger. The metering quantity used may be calculated from the length of the remainder that is to be measured.

To this end, the diameter of the remainder, which, as stated in the foregoing, corresponds to the plunger diameter, the density of the alloy used, and the measured length of the remainder may be used to calculate the cast metering weight. This makes it possible to reliably infer the metering weight, since, of the aforesaid variables, only the length of the remainder may be considered as not constant during a plurality of casting steps and this variable changes due to fluctuations in metering. Since the method for calculating the casting weight uses the length of the remainder, on the one hand it is possible to determine the cast metering quantity without intervening in production, and on the other hand it is possible to reliably infer the cast metering weight, since the remaining press length represents a measure for the casting weight and the casting weight is calculated with constant parameters or with parameters that change only slightly.

The regulating cycle may be constructed such that the remaining press length is captured as a measure for the casting weight or the calculated casting weight as the actual value and this value is compared to the mean value of a tolerance range as the target value. The mean value of the tolerance range may be defined as the mean value between the fixed upper and lower limits of the tolerance range. If a difference between these two values exceeds a predetermined, adjustable value, the adjustable metering weight used as correcting variable for the regulating cycle is revised so that the actual value remains in the middle of the tolerance range, that is, near the target value, and the control difference is minimized. This means that if the remaining press length is too long, that is, if the prior metering weight is set too high, the metering weight for the next casting process is reduce, while it is correspondingly increased if the prior metering weight was set too low. The set metering weight should only be revised, however, if a difference, which defines the control difference, between the remaining press length or the variable derived therefrom and the mean value of the tolerance range exceeds a predetermined maximum value. A setting parameter of the die casting cell that influences metering weight may be a pressure-time integral in a metering furnace, for instance.

Some statistical metering fluctuations are corrected by using a mean value of the remaining press lengths or of the casting weights and/or by using integral calculation methods, since the weights of the castings always distribute in a certain range. This distribution is taken into account by a tolerance range for the remaining press lengths or the casting weights. In contrast, an immediate revision of the metering weight after the determination of individual remaining press lengths or casting weights may lead to undesired fluctuations in the set metering weight. In addition to compensating statistical deviations by using a mean value and/or integral calculation methods, systematic deviations are also corrected using the regulating cycle. The revision of the metering weights between individual casting processes occurs in small steps in order to operate the machine more uniformly and to prevent the jumps in metering weight mentioned in the foregoing.

In the regulating cycle, the formed mean value is taken for this as actual value adjusted by distribution. This method offers the advantage that an actual value in the form of the mean value is used that may be guided by setting the target value of the provided metering weight in the tolerance range and is corrected by drifts that result from the systematic fluctuations in the metering quantity. In this manner compensation of both statistical and systematic metering fluctuations is achieved and thus the number of rejected parts is reduced.

One advantageous refinement provides that the mean value or the integral of the remaining press lengths or of the casting weights is not calculated until after a predetermined minimum number of measurements. Thus it is assured that the required number of measurements must be available before a statistical measure like the mean value or the integral may be reasonably calculated and thus the influence of the distribution is reduced. If the number of measured values is too low, the distribution is generally greater, which may lead to undesired compensation during the regulation.

In order to keep the determination of the mean value or integral more precise, these may be determined continuously from all measurements or, for instance, only from a predetermined number of measurements. After changes to the machine, for instance cleaning and removal of deposited residual metal, the old values are no longer used for forming the mean value, but rather the evaluation may be limited to a defined number of values.

In addition to the length of the remainder and a therefore indirect determination of the metering weight, other process data may be used for the calculation. These may include, inter alia, supply pressure and supply time or data for the casting mould, such as casting mould temperature, or parameters for the die casting cell, such as the speed of a plunger that presses the melt into the casting mould.

In another advantageous refinement, values for forming the mean value or integral that are outside of a predetermined range, for example a confidence range, are not taken into account. Values that are outside of this interval, if they occur individually, are considered outliers that are not used for the data basis of the regulation algorithm. This may be the case, e.g. during unusual processes such as clamped valves for supplying the supply pressure, incorrect metal transfer between metering furnace and casting machine, freezing of some of the melt in a channel that is too cold, deposits on the bottom of the metering furnace that block the metering pipe, or the like. However, this also includes the less frequently occurring mould injectors, i.e. castings in which metal melt escapes during the pressing process due to a mould that is not tight.

During systematic interventions in the machine and casting process, for instance cleaning or refilling the furnace, in order for there not to be any account taken in the mean value formation of the associated metering corrections, such interventions and metering corrections are treated as disturbance variables and are automatically reregulated after the intervention. Refill processes may be detected automatically by measuring the fill level of the melt. For detecting cleaning processes, confirmation of such an intervention by the user of the die casting cell is required, for instance by pressing a button provided for this purpose .

During a furnace cleaning, as a systematic metering correction the metering weight may be automatically reregulated during and after the correction.

During other systematic metering corrections, such as refilling melt, in addition to regulating the metering weight, it is also possible to regulate a correction factor for the refilling compensation, the so-called Z factor. Since during a refill compensation the filling speed is a critical parameter and the product of correction factor and filling speed goes into the compensation, regulation of the correction factor offers advantages over merely regulating the metering weight.

Creep-effect compensation following a refilling process represents another systematic metering correction. Here, the set metering weight is reduced over a certain number of metering processes, wherein the reduction grows increasingly smaller. In addition to regulating the metering weight, which naturally includes the described reduction, it is also possible to regulate the number of metering processes to be corrected.

The effect of systematic interventions may advantageously be quantified and evaluated through the determination of the metering weight using the remainder. This includes in particular both the cleaning of the metering pipe and channel region and the compensation of refilling processes and the so-called creep effects of the first metering processes after refilling. These creep effects are caused by the residual quantity in the furnace mixing with newly added melt. The evaluation of one or a plurality of similar systematic interventions may advantageously be applied, by mean formation or by integral calculation methods, to the formation of new correction values or to the improvement of existing correction values in that these correction values are revised upward or downward by comparing the associated metering weights during a systematic intervention to the target value (mean of the tolerance range).

The method is used particularly advantageously in aluminum pressure die casting.

One inventive device comprises a die casting cell having a casting mould, a press for pressing the metal into the casting mould, and a metering device for filling the press with a guantity of metal which is determined by the metering weight, and further comprises a measuring unit and a regulating unit. The object of the measuring unit is to measure the length of the remainder continuously and to transmit this measured value to the regulating unit. The regulation then occurs thereafter, depending on the remaining press length, through the regulating unit according to the method described in the foregoing. The metering device may be a metering furnace or a scoop system.

In an advantageous refinement, the regulating unit is designed in such a way as to calculate the casting weight from the length of the remainder and predetermined parameters, such as e.g. the plunger diameter and the density of the alloy used. The regulation then occurs, depending on the casting weight, through the regulating unit according to the method described in the foregoing.

The regulating unit is advantageously designed in such a way that, after a plurality of measurements, the mean of the remaining press length or of the casting weights or an integral is calculated using the measured values of the remaining press length or of the casting weights and the regulation is conducted such that the mean value or the integral of the remaining press length or of the casting weights is in the middle of a predetermined permissible tolerance range .

One advantageous refinement of the device provides that an integrating regulator (I regulator) having adjustable damping constants is present in order to adjust the regulation precisely but slowly. In this way, especially when using the mean value and/or by applying integral calculation methods, great jumps in the set metering weight are prevented.

In one advantageous refinement, the damping constant is adjustable in such a way that there is no need to form a mean value. In this case, the intervention of the regulator occurs correspondingly slowly over a number of steps so that the same result is achieved as with the more complicated mean value formation.

Exemplary embodiments of the invention are illustrated in the drawings and are explained in the following using Figures 1 through 6.

Fig. 1 Cross-section of a casting mould for a die casting cell and metering device having measuring and regulating device;

Fig. 2 A flow-chart for a regulating method;

Fig. 3 A flow-chart for a regulating method having an integral element with mean value for- mation;

Fig. 4 A flow-chart for a regulating method having an integral element without mean value formation;

Fig. 5 A flow-chart for a regulating method during systematic metering corrections;

Fig. 6 A schematic representation of a regulation of the set metering weights depending on the actual metering weights of the cast castings .

Fig. 1 depicts a cross-section of a die casting cell for aluminum pressure die casting having a casting mould 1, a plunger 2, a casting 3, and a remainder 4. The remaining press length 5 marked by the double arrow is measured by a schematically illustrated measuring unit 6 and transmitted to a regulating unit 7. This regulating unit 7 is connected to a metering device 8, in this case a metering furnace, that conducts the melt into the shot sleeve, where it is pressed into the casting mould 1 by the plunger 2. A different metering device 8, e.g. a scoop system, may also be used instead of a metering furnace. Once the metal has hardened into the casting 3, the latter may be removed from the casting mould 1. After the pressure die casting has concluded, a sprue of the casting 3, which comprises the so-called remainder 4, remains on the casting 3 and the remainder 4 is removed therefrom before the next casting step is conducted. The length of this remainder 4, the remaining press length 5, is measured and used to calculate the metering weight. The metering weight is calculated with a formula:

DGaktueii is the metering weight of the most recently cast casting 3, DGm±ttei is the mean metering weight, wherein a mean remainder having the length PRmittei is produced with this metering weight, PR-aktueii is the remaining press length of the most recently cast casting 3, PRmittei is the remaining press length that results from the mean of the upper and lower limits of the tolerance range, d is the plunger diameter, and p is the density of the metal used. In addition to the given method for calculating the metering weight, the metering weight may also result using an evaluation of a combination of measured remainder data and other process data, such as e.g. special casting mould temperatures. If the previous remaining press length 5 is too short, the metering weight for the next casting steps is increased, if the remaining press length 5 is too long, the metering weight is correspondingly reduced. The metal used for executing the method comprises aluminum.

Fig. 2 depicts a flow-chart of the regulating method. The regulation begins with the measurement of the remaining press length 5. Then the weight of a casting 3 is calculated from the obtained measured value and other parameters that are considered constant. In the next step there is a check of whether the calculated weight is in a predetermined confidence interval. If not, there must be a check of whether the specific value derives from a systematic metering correction, such as a cleaning of the furnace. If this is the case, a revision that is explained in greater detail in Fig. 5 is undertaken.

However, if the value does not derive from a systematic metering correction, it must be checked whether values immediately preceding the current value were also outside of the confidence interval. If this is true, the metering weight is revised in a plurality of casting steps, since the measured values indicate a systematic deviation and the target value must be revised to reduce the number of rejects. On the other hand, if only a single value is outside of the confidence interval, the system considers it an outlier and the target value is not revised. In order not to falsify mean value formation, the outlier is also not used for calculating a mean value of the casting weights 10 in the following casting steps.

If the calculated casting weight is in the confidence interval, a mean value 10 as the actual value of the control circuit is formed from the current casting weight and already determined casting weights. This mean value 10 may only be calculated after a predetermined minimum number of measurements, but it may also be formed continuously from all measurements or from only a selected number of measurements. From this actual value, the difference to the mean value of the tolerance range, which is established by the upper and lower limits 9 of the range, and its amount is calculated. If the amount of the difference is less than a defined value, the casting process is conducted without changing the metering weight.

If the amount of the difference exceeds a defined value, however, there is a further check of whether a certain number of amounts of prior differences also exceed this value. If this is not the case, there is a further check of whether the algebraic signs of the prior differences are the same and are identical to the algebraic sign of the currently occurring difference. If this is the case, the metering weight is revised in a plurality of casting steps, since in this case there is a systematic drift in one direction, otherwise the set metering weight for the casting process is used, since there is no clear drift identified in one direction. The revision of the metering weight occurs here and in the subsequent exemplary embodiments by adjusting a set parameter, for instance using a pressure-time integral over a casting process .

If the amounts of the prior differences do not exceed the amount of the currently determined difference between actual value and mean value of the tolerance range, the casting process should likewise be started with the set value.

The metering weight should be revised over a plurality of casting steps. To this end, first a correction variable, by which the metering weight is to be changed in the next casting steps, is formed by dividing the difference between actual value and mean value of the tolerance range by the number of desired steps over which a revision is to occur. However, since correction variables from prior revisions must also still be taken into account, these are calculated arithmetically and the relevant correction variable is thus retained in the casting step. In addition, it is checked whether the correction variable obtained in this manner is small enough, i.e., whether the increment does not exceed a certain value. Otherwise the increment is set to the maximum permissible value and the number of steps over which the revision is to occur is automatically increased correspondingly.

The device used for regulating the method may be a computer that is connected to the die casting cell via an interface. Alternatively, the device for regulating may also be coupled directly to the control unit of the die casting cell, however.

Fig. 3 is a schematic depiction of a flow-chart for the regulating method having an integral element and mean value formation. The difference from the method depicted in Fig. 2, which uses only the mean value formation, is comprised in the use of an integral element. After the difference between the actual value, calculated as the mean value 10, and the mean value of the tolerance range is found, the difference is processed in the integral element. The correction occurs in such a manner that the difference is multiplied by a damping factor 1/D and is added to the difference determined in the previous step:

DosDiff(i) is the recursively defined difference between target value and actual value that is taken for setting the metering weight, while Diff represents the currently determined difference. The damping factor 1/D then is the so-called reset time 1/TN in a classic integrating regulator. The metering weight to be set is then calculated recursively using the following formula:

DosEinstell a) is the parameter for the metering weight to be set. Incremental correction occurs in this integral method using the damping factor 1/D, which only slightly weights the target-actual differences when selected. The difference is added up and the intervention becomes greater only if there are differences with the same algebraic sign over a plurality of steps. When there are statistical fluctuations about a mean value, these are compensated in the sum and there is no intervention. By using integral element and mean value formation, statistical errors in the regulating method are taken into account in a special manner.

An alternative to the method described in Fig. 3 is depicted in Fig. 4, in which an integrating controller is used but there is no mean value formation. In this case, it is necessary to select the damping factor correspondingly small, that is, to select the variable D correspondingly large. This results in a slower intervention by the regulator, which has an effect similar to when the mean value is taken into account, but is simpler to realize.

Fig. 5 depicts a regulating method for systematic metering corrections as a flow-chart. If a systematic metering correction, such as a cleaning process, is acknowledged by the user, the mean increase in the set metering weight is determined by calculating a best fit line through the series of set metering weights between the last and the current cleaning process. Following the current cleaning process, this mean increase in the current metering weight is subtracted from the currently set metering weight in order to satisfy the figures for a cleaned metering pipe. A refill process is not acknowledged by the user, but instead is established by measuring the hydrostatic increase in pressure in the furnace melt, which overlays the furnace inner pressure. The increase in pressure causes an increased mass flow of metered melt, which is corrected by shortening the metering time or reducing the pressure integral. Associated with this is sampling the filling speed, that is the fill level increase per unit of time. According to the prior art, such refilling compensation is calculated as follows:

Z is the correction factor to be set for the refill compensation, by which factor the filling speed is multiplied. This factor is determined empirically for each furnace and optimized. When there is refilling, using the inventive method makes it possible not only to revise the metering weight automatically, but also to revise the correction factor automatically.

Creep-effect correction corrects the set metering weight for the metering processes following refilling. The set metering weight is reduced in increasingly smaller steps over a certain number of metering processes, for instance according to the following formula:

n indicates the number of metering processes over which a correction is made, i is the continuous index, which runs from 1 to n, kg Minderung is the set value for a reduction in the metering weight, DosGewNachwirkungsk0rrektur snd DosGewe±ngesteiit indicate the metering weight to be set for a creep-effect correction or the set metering weight. When performing an inventive method with determination of the remainder, therefore, the correction values n and kg Minderung may be optimized automatically instead of varying the set metering weight directly.

Fig. 6 is a schematic depiction of regulation of the set metering weights depending on the actual metering weights of the cast castings 3. Fig. 6 a) depicts the course of the set metering weights when conducting the regulation, while Fig. 6 b) illustrates the weights of the cast castings 3 calculated from the remainder 4 and the limits 9 of the tolerance range. The actual weight of the castings 3, which was calculated from the remainder 4, fluctuates within the upper and lower limits 9. The mean value 10 of the actual casting weights is maintained in the middle of the tolerance range by regulating the metering weights. In the situation illustrated in Fig. 6, no rejects are produced since all of the casting weights may be kept within the limits 9 of the tolerance range. The differences in weight between set metering weight and calculated weight result for instance due to metal that deposits in the metering pipe or that remains in the channel. Although this metal has been metered, it has not travelled into the shot sleeve. Oxide deposits on the lip of the metering pipe change the hydrostatic reference pressure level and also influence the metering weight. It would also be possible for a new metering tube with different dimensions to be used or for newly added melt and melt that has remained in the system to be mixed, which changes the properties of the melt.

List of reference signs: 1 Casting mould 2 Plunger 3 Casting 4 Remainder 5 Remaining press length 6 Measuring unit 7 Regulating unit 8 Metering device 9 Limit of tolerance range 10 Mean value of casting weights (actual value)

Claims (14)

A method of regulating a metal dosage by producing a plurality of mold parts (3) with a press molding cell, wherein a quantity of liquid metal, determined in each case by its dosing weight, is introduced into a filling chamber and pressed into a mold ( 1) with a press plunger (2) and in the mold (1) solidify to one of the molding parts (3), the method comprising regulating a press residue length (5) by the following steps: a) continuously measuring the press residue length (5) as the length of a pressing residue (4) of the molded parts (3) made; b) determining a control difference by comparing the press residue length (5) or a derived size with a mean of a tolerance range; c) setting a setting parameter of the press molding cell affecting the dosing weight, depending on the established regulatory difference, in such a way as to minimize the regulatory difference, where a dosage amount change is automatically compensated resulting from a predetermined systematic dosing correction. Method according to claim 1, characterized in that the set dosing weight for controlling the press residue length (5) is corrected only if a difference defining the control difference between the press residue length (5) or the derived value and the tolerance range mean value exceeds a predetermined maximum value. Method according to one of claims 1 or 2, characterized in that the derived size used to determine the control difference is given by the compressor residual length (5) calculated on average for or integrated over several mold parts (3). Method according to one of Claims 1 or 2, characterized in that the derived size is given by casting weight or a casting weight calculated on average for or integrated over several casting parts (3), wherein the casting weight is calculated from the pressing residual length (5) and predetermined parameters. Method according to one of Claims 3 or 4, characterized in that an average value (10) or an integral formed for determining the derivative size is calculated continuously over all casting parts (3) or in each case over a predetermined amount. number of previously made casting parts (3). Method according to claim 5, characterized in that, when calculating the mean (10) or integral, values which are outside a predetermined range, of the pressing residue or the size derived therefrom are not taken into account. Method according to one of the preceding claims, characterized in that the setting of the setting parameter is based on an analysis of a combination of measured press residue lengths (5) and further process data. Method according to one of the preceding claims, characterized in that the setting parameter is a pressure-time integral defined by a dosing process in a dosing oven. Method according to claim 1, characterized in that the systematic dosing correction is carried out by an oven cleaning or by refilling of the melt. Apparatus for carrying out the method according to one of claims 1 to 9, comprising a molding cell with a mold (1), a press for pressing the metal into the mold (1) and a metering device (8) for filling the amount a metal determined by the metering weight in the press, further comprising a measuring unit (6) and a control unit (7), the measuring unit (6) being designed to continuously measure the length of the press residue (4) and transfer it to the control unit ( 7), and wherein the control unit (7) is designed to determine the control difference by comparing the press residue (5) or a derived size with the mean of a tolerance range and setting a setting parameter of the press molding cell affecting the dosing weight, depending on the determined regulatory difference. such that the control differential is minimized where the device is designed to automatically compensate for a dosage quantity change resulting from a predetermined systematic dose correction. Device according to claim 10, characterized in that the metering device (8) is a metering oven or spoon system. Device according to one of claims 10 or 11, characterized in that the regulating unit (7) is designed to calculate the casting weight from the length and predetermined parameters of the press residue (4) and that the dosing weight can be adjusted depending on the calculated casting weight by the regulating unit. (7) in such a way that the casting part weight lies in the middle of a predetermined tolerance range. Device according to one of claims 10 to 12, characterized in that the control unit (7) is designed to calculate a mean value (10) or integral from a plurality of press residues (4) or casting weights, and that the dosage weight can be adjusted depending on of the formed mean (10) or integral by means of the control unit (7) in such a way that the mean (10) or integral of the casting part weights lies in the middle in a predetermined permitted tolerance range. Device according to one of claims 10 to 13, characterized in that the control unit (7) comprises an integrating regulator with adjustable damping constant for controlling the dosing weight.