AT521579B1 - Method for detecting inhomogeneities in melts - Google Patents

Abstract

A method for detecting inhomogeneities in melts, in particular plastic melts, wherein at least one first ultrasonic signal is transmitted in such a way that it passes at least partially through a first melt volume, the at least one first ultrasonic signal that has passed through the first melt volume is measured while generating at least one first measurement signal, at least a second ultrasonic signal is emitted in such a way that it passes at least partially through a second melt volume different from the first melt volume, the at least one second ultrasonic signal which has passed through the second melt volume is measured with the generation of at least one second measurement signal, the at least one first measurement signal and the at least a second measurement signal with regard to an amplitude and / or a variable that can be mathematically derived therefrom, on the one hand, and / or a transit time and / or a variable that can be mathematically derived therefrom, on the other hand and a distinction is made as inhomogeneities in the melt on the basis of the comparison between foreign objects and incompletely melted or plasticized partial volumes, whereby on the basis of the comparison of the at least one first measurement signal and the at least one second measurement signal with regard to the amplitude and / or a computationally derivable one Size on the one hand and the running time and / or a value that can be calculated therefrom on the other hand, a distinction is made between different types of foreign objects and incompletely melted or plasticized partial volumes as inhomogeneities in the melt.

Description

description

The present invention relates to a method for detecting inhomogeneities in melts, in particular plastic melts, according to the features of the preamble of claim 1 and a method for detecting deviations from state variables of melts in a molding process carried out in molding cycles. The invention also relates to the use of a plasticizing unit for a molding machine and / or a molding machine.

Under molding machines, for example, injection molding machines, transfer presses, presses, casting machines, die casting machines and the like can be understood. Accordingly, molding processes can be understood to mean, for example, injection molding processes, transfer molding processes, pressing processes, casting processes, die casting processes and the like.

In the following, experiences and test results with regard to injection molding processes and plastic melts are described which the applicant - unless explicitly stated otherwise - is available internally (internal state of the art). These experiences and test results can, however, be transferred to general shaping processes.

In injection molding machines, the plastic melt is usually plasticized in a plasticizing unit. As standard, a plasticizing unit consists of a heated steel cylinder (plasticizing cylinder) in which a rotating screw (plasticizing screw) conveys the - mostly granular - plastic raw material in the direction of the screw antechamber, compressing, shearing and thereby melting it. The screw antechamber serves as a store for the plasticized material and is enlarged by the axial retraction of the screw during the plasticizing process.

The withdrawal of the screw leads to a shortening of the active or effective screw length and thus to a decrease in the melt temperature. If this drop in temperature and the conveying speed of the plastic material are too high, inhomogeneities in the form of non-plasticized granules can arise in the melt. Incorrectly set plasticizing parameters (e.g. speed, cylinder temperature, back pressure) and / or moist base material as well as other chemical processes - can also lead to gas bubbles forming in the melt. Other inclusions in the melt can be, for example, metallic wear particles from the tribological system of mass cylinder-non-return valve-plasticizing screw, contaminated base material or thermally degraded plastic material. The injection (by the piston-like axial forward movement of the plasticizing screw) of a melt contaminated with structural inhomogeneities in this way into the cavities of an injection molding tool leads to impairment of the optical and / or mechanical properties of the molded part or can interrupt the production cycle in that contaminants close the sprue system.

A similar problem to inhomogeneities is the change in thermodynamic state variables of the plasticized melt between individual production cycles. Due to the requirements of the real plasticizing process, it is impossible to produce an ideal thermally homogenized melt and thus there are also inhomogeneities in the thermodynamically coupled variables of pressure , Density and compressibility. However, the goal of a stable manufacturing process must be that the temperature and density of the plasticized material as a function of the radial and axial position in the screw antechamber is as reproducible as possible - one then speaks of a steady thermal process. Such fluctuations from molding cycle to molding cycle - hereinafter also referred to as shot-to-shot inhomogeneities - lead, for example, to weight fluctuations and changes in the shrinkage behavior of the plastic molded parts produced. The main reasons for such fluctuations are start-up processes, process interruptions, batch fluctuations in the plastic raw material and changes in environmental influences.

The inhomogeneities described at the beginning in the form of foreign objects, for example, such as metal particles, gas bubbles and the like and not completely plasticized plastic are referred to as structural inhomogeneities in contrast to the shot-to-shot inhomogeneities mentioned.

Checking whether the selected plasticizing parameters lead to non-plasticized granules or gas bubbles is often carried out in practice by spraying the melt out into the open and visually assessing the melt strand. This method is therefore dependent on the experience of the machine setter (operator) and cannot be classified as reproducible. Other inclusions (metallic wear particles, contamination of the granulate) are only possible by means of a complex visual inspection or machine inspection (e.g. camera systems) after the manufacturing process has taken place.

Changes in the thermodynamic state variables of the plasticized melt can in principle be detected, for example, with temperature measurement (e.g. infrared, thermocouple or ultrasound) or by measuring the weight of the manufactured components. Systems for measuring the temperature in the antechamber using ultrasound measurements are disclosed, for example, in the documents AT 512 647 A4 and AT 516 452 A1, whereby the dependence of the speed of sound on the pressure and temperature of the plastic melt is used. While the temperature measurement requires the corresponding knowledge of the machine operator in order to interpret the data, the weight measurement can only take place after the actual production with the corresponding effort.

In addition, EP 2179835 A1, WO0196854 A2, US 4509360 A, WO8505449 A1, US 2017021548 A1 and JP H0275110 A are known as general prior art. They disclose methods for detecting inhomogeneities.

The object of the present invention is therefore to provide methods which allow inhomogeneities in a melt and / or shot-to-shot inhomogeneities to be detected as reproducibly as possible.

With regard to the structure inhomogeneities, this is achieved by the features of claim 1 by the at least one first measurement signal and the at least one second measurement signal with respect to an amplitude and / or a mathematically derived variable on the one hand and / or a transit time and / or On the other hand, a variable that can be mathematically derived therefrom can be compared and, on the basis of the comparison, a distinction is made between foreign objects and incompletely melted or plasticized partial volumes as inhomogeneities in the melt.

With regard to the shot-to-shot inhomogeneities, the object is achieved by the features of claim 11 by

- a melt is provided for each of the molding cycles,

- a continuous or pulsed ultrasonic signal is sent out during the shaping cycles in such a way that the continuous or pulsed ultrasonic signal passes through the melt,

- the continuous or pulsed ultrasonic signal that has passed through the melt is measured while generating a large number of measurement signals,

a parameter representative of a state variable of the melt is determined from the measurement signals from at least two of the shaping cycles, and

the parameters representative of the state variable for the at least two shaping cycles are compared with one another.

According to the invention, shot-to-shot inhomogeneities can already be detected by comparing measurement signals from at least two of the shaping cycles. The effectiveness of the detection of shot-to-shot inhomogeneities can be improved if measurement signals from a large number of the molding cycles - preferably the measurement signals from all molding cycles - are used.

It can be particularly preferably provided if a plurality of the shaping cycles

len - preferably all molding cycles - are carried out in a system, in particular a molding machine.

The two solutions according to the invention have in common that through the use and comparison of the respective ultrasonic signals applied for the application (structure inhomogeneities, shot-to-shot inhomogeneities), the inhomogeneities corresponding to the application can be classified and recorded - preferably in real time . In comparison, it was only known in the prior art to record temperatures and temperature profiles.

By detecting the structure inhomogeneities and / or the shot-to-shot inhomogeneities according to the invention, it is possible to detect the causes of the inhomogeneities and to remedy them (automatically or by an operator).

As mentioned, the invention can preferably be used in plastic melts. However, use in other melts, such as metallic melts, is also conceivable in principle. For the purposes of the invention, the term melt encompasses all flowable masses which can be given a shape by solidifying and / or hardening in a molding tool. For example, a melt with a load (for example fiber or gas load) can also be examined according to the invention.

Inhomogeneities can also arise from inhomogeneously distributed loading.

In the context of shot-to-shot inhomogeneities, both thermodynamic state variables and other state variables (for example a load distribution) can be examined.

The fact that melt volumes differ can be understood to mean that they are not the same. It is not absolutely necessary that the melt volumes are disjoint. They can therefore overlap.

The expression “ultrasonic signal (s)” is used as a generic term for the at least one first ultrasonic signal, the at least one second ultrasonic signal and the continuous or pulsed ultrasonic signal.

To generate ultrasonic signals, transmitter units or receiver units (also: transmitter and receiver for short) can be used, which are arranged on opposite sides of the melt. But this is not absolutely necessary. For example, integrated transmitting and receiving units can also be used, which receive the transmitted ultrasonic signal, for example after a reflection. The transmitting unit can detect the ultrasonic signal changed by passing through the melt volume or the melt as a measurement signal.

Under from certain variables (for example amplitude or transit time) computationally derivable variables can be understood that by arithmetic (for example. Reciprocal values, squares, integration, differentiation, etc.) and additional information (for example. Geometric information or information about transmitted ultrasonic signals such as frequency, amplitude, etc.,) can be calculated. In the case of a large number of values for the specific variables, maximum and / or minimum values and / or mean values and / or medians or similar operations can also be used for the computational derivation of further variables.

Foreign objects within the meaning of the invention can be (metallic) wear particles, gas bubbles, thermally degraded base material, other impurities in the base material (i.e. before melting or plasticizing) and the like.

In the context of the invention, ultrasonic signals can be transmitted once, several times, continuously or in pulsed fashion. In principle, the invention does not set any limits to the duration and time interval between the pulses during pulsed transmission. However, it is preferred, because it has proven to be technically possible and also useful within certain limits, if several thousand pulses are transmitted per second. The same number of measurement signals can be generated by receiving the pulses.

The at least one first ultrasonic signal and the at least one second ultrasonic signal can be transmitted simultaneously. Of course, there can also be a time interval between the at least one first ultrasonic signal and the at least one second ultrasonic signal. At most, this time interval can be determined by a speed of movement of the melt relative to the transmitting and / or receiving ultrasound device, namely in such a way that the time interval is preferably not so long that inhomogeneities cannot be detected because they are too fast on one Transmitter and / or receiver are moved past. Of course, the at least one first ultrasonic signal and the at least one second ultrasonic signal should be emitted during the same shaping cycle in order to detect structural inhomogeneities.

The detection of structural inhomogeneities in an injection molding process can preferably be carried out during a plasticizing and / or injection process. In principle, however, the at least one first and the at least one second ultrasonic signal can also be emitted during a metering process - or generally during a melting or filling process.

The method according to the invention can particularly preferably be carried out with a plastic melt present in an injection cylinder of an injection molding machine. Of course, the method according to the invention can also be carried out in a plasticizing unit that is separate from the injection unit or in further separate volumes (for example shot pot, hot runner system).

Temperature deviations, pressure deviations, density deviations and the like can be used as deviations in thermodynamic state variables.

As a parameter representative of the state variable, for example, a mean value of values for the thermodynamic state variable calculated from the multiplicity of measurement signals can be used. However, other measures such as a median, maximum or minimum values can also be advantageous in certain situations.

Of course, the solutions according to the invention can also be combined with regard to structural inhomogeneities and shot-to-shot inhomogeneities. That is, the detection of structure inhomogeneities according to the invention can also be carried out in each individual (or any subset) of the shaping cycles carried out. The first melt volume and the second melt volume are then part of the melt for the respective molding cycle.

It is also provided according to the invention that the at least one first measurement signal and the at least one second measurement signal are compared with respect to the amplitude and / or a value that can be mathematically derived therefrom, on the one hand, and the transit time and / or a value that can be calculated therefrom, on the other hand, and based on When comparing different types of foreign objects and incompletely melted or plasticized partial volumes, a distinction is made as inhomogeneities in the melt. By taking into account the transit time and the amplitude (or values derived from these mathematically), in particular the type of foreign objects can be inferred. For example, metallic wear particles in a plastic melt produce a shortened running time (higher speed of sound in the metal compared to a plastic melt) and gas bubbles produce longer running times (lower sound speed in the gas).

Advantageous developments of the invention are defined in the dependent claims.

The possibility of a continuous or pulsed ultrasonic signal has already been mentioned. The continuous or pulsed ultrasonic signal can pass through a plurality of melt volumes, the continuous or pulsed ultrasonic signal at different times and / or in different periods of time preferably being used as the at least one first ultrasonic signal and / or as the at least one second ultrasonic signal. The continuous or pulsed ultrasonic signal that has passed through the multitude of melt volumes can be measured with the generation of a multitude of measurement signals.

By detecting a large number of melt volumes, a particularly reliable detection of structural inhomogeneities can be possible.

When detecting shot-to-shot inhomogeneities, the large number of measurement signals enables the respective shaping cycle to be recorded effectively without “outliers” in individual measurements negatively affecting the detection of shot-to-shot inhomogeneities.

A speed of sound, for example, can be used as a variable that is mathematically derived from the transit time. It can be, for example, an integral or average speed of sound. It can be calculated, for example, using the known sound path through the melt, with additional sound paths, for example through the plasticizing cylinder, being able to be neglected or taken into account. In many cases it is not absolutely necessary to take this into account, as the speed of sound in the plasticizing cylinder, which is usually made of steel, is much higher than in the melt and can therefore only have a very small effect on the measured running time.

The square of the amplitude (intensity), for example, can be used as a variable that is mathematically derived from the amplitude.

The fact that variables such as the transit time and / or the amplitude and / or computationally derived variables differ (or do not differ) can be understood to mean that deviations (e.g. the transit time and / or the amplitude) have a certain predetermined value Reaching or exceeding the limit value (or not reaching or not exceeding).

Inhomogeneities can be recognized by specifying an inhomogeneity limit value for a difference between the at least one first measurement signal and the at least one second measurement signal, and if the inhomogeneity limit value is exceeded, an inhomogeneity is deduced. This can embody the basic concept that changes in the measurement signals (above a certain threshold) indicate the presence of inhomogeneities.

A reference measurement and a current measurement or two successive measurements can be used as the at least one first measurement signal and the at least one second measurement signal.

According to the invention, the types of inhomogeneities can be characterized by changes of varying magnitude in the amplitude or in the magnitude derived therefrom mathematically. In the case of inhomogeneities in the form of not completely melted or plasticized partial volumes, the measurement signals show a more gradual transition, which originates from a more gradual density transition of the not completely melted or plasticized partial volume. Foreign objects are characterized by a very sharp transition in the measurement signals.

Inhomogeneities can therefore preferably be differentiated as follows:

- The presence of foreign objects if a difference between the at least one first measurement signal and the at least one second measurement signal with regard to the amplitude and / or the variable that can be calculated therefrom exceeds a predetermined foreign object limit value, or

- Otherwise, partial volumes that are not completely melted or plasticized are present in the melt.

If the at least one first measurement signal and the at least one second measurement signal are not recorded individually, but a large number of measurement signals are recorded, for example as part of a continuously or pulsed ultrasonic signal, the changes between the individual measurement signals can be interpreted as the slope of a recorded curve will. The use of an inhomogeneity limit value and / or a foreign object limit value can then also be understood as the use of slope ranges as follows: Slope range below the inhomogeneity limit value: no inhomogeneity

- Gradient range above the inhomogeneity limit value and below the foreign object limit value: not completely melted or plasticized partial volume in the melt

- Incline range above the foreign object limit value: foreign object

Further distinguishing criteria can result from the point in time in a molding cycle at which the inhomogeneity occurs. In an injection molding process, for example, plastic that is not completely melted occurs more towards the end of an injection process because of the shorter dwell time in the plasticizing cylinder. Foreign objects such as wear particles, on the other hand, can occur during the entire molding cycle.

Inhomogeneities that occur at the beginning of a molding cycle can therefore be classified as foreign objects with a high degree of probability.

It can be provided that the first melt volume and / or the second melt volume is provided in a screw antechamber arranged in a plasticizing cylinder, with a plasticizing screw arranged in the plasticizing cylinder (also for short below) for producing the first melt volume and / or the second plastic melt volume referred to as a screw) is used. The first melt volume and the second melt volume can then also be referred to as the first plastic melt volume and the second plastic melt volume, respectively.

A preferred in relation to the structural design of the invention, simple solution can be provided by using at least one transmitting unit to transmit the at least one first ultrasonic signal and / or the at least one second ultrasonic signal and that for detecting the at least one first measurement signal and / or at least one receiving unit is used for the at least one second measurement signal, the at least one transmitting unit and the at least one receiving unit preferably being arranged on opposite sides of an injection nozzle mounted on the plasticizing cylinder.

It can be provided that a melt for such transmission of the at least one first ultrasonic signal and the at least one second ultrasonic signal pass through the first melt volume and the second melt volume, respectively, to at least one transmission unit for transmitting the at least one first ultrasonic signal and / or the at least one second ultrasonic signal and / or is moved past at least one receiving unit for detecting the at least one first measurement signal and / or the at least one second measurement signal. In particular in the context of an injection molding process in an embodiment using a plasticizing screw in a plasticizing cylinder, this can be done particularly easily during the injection of the plastic melt into a molding tool, with the melt having to be moved in any case.

Of course, the possibility is in principle also conceivable, for example, to move a pair of transmitting and receiving units with respect to the melt (along an axis and / or in the alignment) or to use several transmitting and / or receiving units.

It can be provided that the parameters representative of the state variable are compared with one another with regard to a tolerance range and preferably molded parts that were produced in those molding cycles whose parameters representative of the state variable are outside the tolerance range are declared as rejects.

The tolerance range can be designed as a predefined value range around a parameter representative of the state variable which was calculated from the measurement signals of one of the at least two shaping cycles. The shaping cycle used in this way to define the tolerance range can then also be referred to as the reference shaping cycle.

In particular when using measurement signals from a large number of shaping cycles, the tolerance range can also be determined by determining a mean value and a standard deviation of parameters representative of the state variable from a large number of

Forming cycles are determined. The tolerance range can then be given by that range which results from the mean value and deviations less than (or equal to) the standard deviation thereof. Instead of the mean value and the standard deviation, it is of course also possible to use similar statistical values, such as a median and different confidence bands.

Independently of this, the tolerance range can of course be determined by calculations or simulations.

In addition to the declaration of scrap molded parts that occur during production in a steady state - quasi as "outliers" - the detection of state variables according to the invention can also determine when the steady state is present and when it is not. As a result, at the start of production or after production interruptions, it can be recognized with a high degree of certainty when good parts are to be expected.

Provision can particularly preferably be made on the basis of the detected foreign objects and / or partial volumes that are not completely melted or plasticized and / or deviations in thermodynamic state variables

- a notice is issued to an operator and / or

- an automatic adjustment of a machine setting is carried out.

To avoid inhomogeneities in the future, the machine settings can be changed. For example, in the case of partial volumes of plastic melt that have not been fully plasticized, a screw speed can be changed (automatically or on the basis of an operator input). In most cases it is necessary to reduce the screw speed. Other measures that can be taken in this case would be, for example, increasing a back pressure in the plasticizing cylinder and / or increasing a temperature of the plasticizing cylinder (for example by increasing a heating power).

Further advantages and details of the invention emerge from the figures and the associated description of the figures. Show:

1 shows a schematic sectional illustration of a screw antechamber to illustrate structural inhomogeneities as they occur in the prior art,

2a and 2b show a first embodiment of the inventive use of ultrasonic transmitters and receivers in an arrangement according to FIG. 1,

3 shows a further embodiment of the use according to the invention of ultrasonic transmitters and receivers in an arrangement according to FIG. 1,

4a and 45 are schematic representations and diagrams to illustrate measuring principles according to the invention for detecting structural inhomogeneities,

5 shows an example of amplitudes and sound velocities when a structure inhomogeneity occurs in a shaping cycle as well

6 shows two diagrams to illustrate a measuring principle according to the invention for detecting shot-to-shot inhomogeneities.

One embodiment of the invention includes the evaluation of ultrasonic signals which are sent by a plastic melt during the injection process in an injection molding process. The evaluation of the signals can be used to detect inhomogeneities in the melt or deviations from the thermally steady state. After the problem has been classified automatically, suggestions can be made to the machine operator as to how he can eliminate the problem by changing machine setting parameters or, alternatively, a fully automated process (e.g. implemented in the machine control) can try to eliminate the classified problem independently by changing machine setting parameters.

Fig. 1 shows the schematic representation of a plasticizing unit consisting of the mass cylinder 1 (plasticizing cylinder), flange 2, nozzle 3, screw 4 (plasticizing screw) and various heating bands 5. In front of the plasticizing screw 4 is the plastic melt to be examined after the plasticizing 6. This is largely injected into the injection mold. Inhomogeneities 7 can be found in this plastic melt 6.

2a shows the plasticizing unit during the injection process. The inhomogeneities 7 are conveyed with the melt 6 by the forward movement of the screw 4 over the flange 2 and the nozzle 3 in the direction of the tool. If one now sends ultrasonic pulses with a correspondingly high pulse repetition frequency through this flowing melt, inhomogeneities 7 can be detected. For this purpose, the amplitude and transit time of the ultrasonic pulses are recorded and evaluated. A possible temporary sensor arrangement is shown in FIGS. 2a and 2b. The sensors (transmitter 8 and receiver 9) are attached to the nozzle 3 opposite one another. The sensors should be attached in such a way that the measuring axis 10 is in the middle of the melt channel, since tests have shown that inhomogeneities 7 are mainly promoted in the middle of the flow channel due to the prevailing flow conditions. It is particularly advantageous to use two opposing surfaces of the hexagon nut of the nozzle (see FIG. 2 b), but other positions on the nozzle, the flange or the mass cylinder can also be used. This attachment can be carried out temporarily (e.g. by means of magnetic holders, pressing or gluing) and is used to optimize the process settings when the process is started.

3 shows an example arrangement for a permanent measurement. The ultrasonic sensors (sensor 8 and receiver 9) are permanently installed on a feed path 11 with a thread and can, for example, be screwed into opposing sensor bores in flange 2. Such sensor bores can of course also be provided in the nozzle 3 or on the mass cylinder 1, or they can be attached in a measuring flange between the nozzle 3 and the flange 2.

4a shows an example of the change in the transit times and amplitudes for different classes of inhomogeneities (F2, F3) compared to a reference measurement without inhomogeneities (F1). The ultrasonic sensors (transmitter 8, receiver 9) are attached to the outer wall of a steel cylinder 12 (for example formed by the plasticizing cylinder 1 or the nozzle 3). The melt 6 is located in a bore in the steel cylinder 12. In the melt 6, the inhomogeneities are located centrally in the measuring axis 10 of the respective transmitter-receiver pair. The melt 6 has a homogeneous temperature T and pressure P. In the case without inhomogeneity F1, depending on the type of polymer melt, there is a certain damping of the sound pulse when passing through the melt (the damping in the steel cylinder is to be regarded as constant and negligible), the running time of the sound pulse depends on the density and compressibility of the melt (the transit time in the steel cylinder is almost constant and can be subtracted from the total transit time of the ultrasonic pulse). The resulting amplitude A1 can be used as the reference amplitude Aref. The resulting transit time t1 can be used as the reference transit time ter (from which an integral speed of sound Vrer can be calculated if the sound path is known). A distinction is made between two types of inhomogeneities: Foreign objects F2 are inhomogeneities with a defined constant density boundary surface (e.g. metal particles or gas bubbles) and incompletely plasticized partial melt volumes F3 are inhomogeneities with a mainly flowing density profile (e.g. unmelted granulate). In the case of an inhomogeneity of type F2, depending on the size and density of the inhomogeneity, in addition to the scattering, there are strong reflection losses of the ultrasound. The speed of sound usually remains almost constant (with the typical relationships between the total sound path through the melt and the size of the inhomogeneity in injection molding). In the case of only partially melted granules, the mainly flowing density profile (density increases into the interior of the inhomogeneity) usually does not lead to a strong amplitude reduction as in the case of inhomogeneities of type F2. However, unmelted granules can best be detected or classified by the transit time or the integral speed of sound in the melt channel: The transit time is reduced or the integral speed of sound

The density increases according to the mean density of the inhomogeneity.

4b shows the various types of inhomogeneity in an alternative (and in many cases more practical) type of representation with graphs for the amplitude and the integral speed of sound.

Purely by way of example, the melt is not guided past the measuring axis 10 of the transmitter-receiver pair, but rather the transmitter / receiver pair is moved along the outer wall at a constant speed v and the amplitude and the integral speed of sound are determined as a function of the measurement position in the melt channel .

While the inhomogeneity of type F2 results in a relatively constant reduction in the pulse amplitude, inhomogeneities of type F3 lead to a gradual change in the pulse amplitude.

The size of the inhomogeneities can be inferred from the width of the amplitude reduction. With the integral speed of sound, type F2 does not show any inhomogeneities, an increased (density of inhomogeneity is greater than melt density, e.g. wear particles from steel) or a reduced mean sound speed (density of inhomogeneity is less than melt density, e.g. gas bubble), which can be seen in the figure 4b is represented by three alternative curves of the integral speed of sound in the area of the inhomogeneity of type F2.

Unmelted granules - that is to say inhomogeneities of type F3 - can in turn be clearly classified by the integral sound velocity in the melt channel running gradually (corresponding to the density distribution in the inhomogeneity).

During the actual measurement, the transmitter / receiver pair is fixed in one position (see FIGS. 2a and 3) and the melt (with any inhomogeneities that may be present) is pushed through the measurement axis at an almost constant speed. Due to the high pulse repetition frequencies, you get a quasi-continuous result of the amplitude and the transit time (or the integral speed of sound in the melt channel derived from it) of the ultrasonic pulse as a function of the measurement duration, which typically corresponds to the injection duration.

5 shows the typical profile of the received amplitude and the integral speed of sound in the melt channel as a function of the measurement duration. The increase in pressure at the beginning of the injection phase leads to an increase in the integral speed of sound and the sound amplitude. The course during the injection phase is relatively constant for both parameters (depending on the axial temperature gradient in the screw antechamber). The so-called holding pressure phase follows after the injection phase. The holding pressure is usually lower than the injection pressure, and both measured variables are reduced accordingly. At the end of the holding pressure phase, both measured variables drop to their initial level. The effect of an inhomogeneity (unmelted granulate) on the measurement results is also shown as an example.

In addition to discrete structural inhomogeneities in the melt, the constancy of the thermodynamic state variables in the melt can also be monitored on a cycle-by-cycle basis. For this purpose, it is proposed to evaluate the mean speed of sound in the injection phase in a suitable manner. This can happen by determining an average value of the quasi-continuous integral sound velocities of the injection phase. Alternatively, the maximum or median value can also be used, for example.

To illustrate the detection of such shot-to-shot inhomogeneities, an experimental result is shown in FIG. 6. Starting from a thermally steady state, the target cylinder temperature for cycle number 11 is set from Tzy = 180 ° C to Tzw = 200 ° C without interrupting component production. The heating of the cylinder leads to a warmer plastic melt and thus to a lower density. As a result, a decrease in the weight of the manufactured components can be seen. This can also be based on the decrease in the mean value of the integral speed of sound during the injection process.

clearly identified - even before the component is demolded. The weight of the component (and, in parallel, the mean integral speed of sound) decrease rapidly until the target cylinder temperature is reached (cycle 19). In this example, it then takes another 6 cycles for the process to stabilize thermally. From cycle 25, the component weights are constant again, which can also be seen from the mean integral speed of sound.

Claims (15)

Claims 1. A method for detecting inhomogeneities in melts, in particular plastic melts, wherein - at least one first ultrasonic signal is transmitted in such a way that it at least partially passes through a first melt volume, - the at least one first ultrasonic signal that has passed through the first melt volume is measured while generating at least one first measurement signal, - At least one second ultrasonic signal is emitted in such a way that it passes at least partially through a second melt volume that is different from the first melt volume, - the at least one second ultrasonic signal that has passed through the second melt volume is measured while generating at least one second measurement signal, characterized in that the at least one first measurement signal and the at least a second measurement signal with regard to an amplitude and / or a variable that can be mathematically derived therefrom, on the one hand, and a transit time and / or one that can be mathematically derived therefrom Size can be compared on the other hand and that on the basis of the comparison of at least a first measurement signal and the at least one second measurement signal with regard to the Amplitude and / or a value that can be mathematically derived therefrom, on the one hand, and the transit time and / or a variable that can be mathematically derived therefrom, on the other hand, between different Types of foreign objects and not fully melted or plasticized partial volumes men is differentiated as inhomogeneities in the melt. 2. The method according to claim 1, characterized in that a continuous or pulsed ultrasonic signal is transmitted in such a way that the continuous or pulsed ultrasonic signal passes through a large number of melt volumes, preferably as the at least one first ultrasonic signal and / or as the at least one second ultrasonic signal the continuous or pulsed ultrasonic signal is used at different times and / or in different time periods. 3. The method according to claim 2, characterized in that the continuous or pulsed ultrasonic signal which has passed through the plurality of melt volumes is measured while generating a plurality of measurement signals. 4. The method according to any one of the preceding claims, characterized in that an inhomogeneity limit value is specified for a difference between the at least one first measurement signal and the at least one second measurement signal, and an inhomogeneity is concluded when the inhomogeneity limit value is exceeded. 5. The method according to any one of the preceding claims, characterized in that inhomogeneities are distinguished as follows: - The presence of foreign objects if a difference between the at least one first measurement signal and the at least one second measurement signal with regard to the amplitude and / or the variable calculated therefrom exceeds a predetermined foreign object limit value, or - Otherwise, partial volumes that are not completely melted or plasticized are present in the melt. 6. The method according to any one of the preceding claims, characterized in that a speed of sound is used as a variable that is mathematically derived from the transit time. 7. The method according to any one of the preceding claims, characterized in that the first melt volume and / or the second melt volume is used in a shaping process - preferably carried out in cycles. 8. The method according to any one of the preceding claims, characterized in that the melt is a plastic melt, wherein the first melt volume and / or the second melt volume is provided in a screw antechamber arranged in a plasticizing cylinder and for producing the first melt volume and / or the second melt volume a plasticizing screw arranged in the plasticizing cylinder is used. 9. The method according to any one of the preceding claims, characterized in that at least one transmitting unit is used to transmit the at least one first ultrasonic signal and / or the at least one second ultrasonic signal and that for detecting the at least one first measurement signal and / or the at least one second measurement signal at least one receiving unit is used, wherein the at least one transmitting unit and the at least one receiving unit are preferably arranged on opposite sides of an injection nozzle mounted on the plasticizing cylinder. 10. The method according to any one of the preceding claims, characterized in that a melt for transmitting the at least one first ultrasonic signal and the at least one second ultrasonic signal that they pass through the first melt volume and the second melt volume, to at least one transmission unit for sending the at least a first ultrasonic signal and / or the at least one second ultrasonic signal and / or at least one receiving unit for detecting the at least one first measurement signal and / or the at least one second measurement signal is moved past. 11. The method according to any one of the preceding claims, for detecting deviations from state variables of melts in a molding process carried out in molding cycles, wherein - a melt is provided for each of the molding cycles, - a continuous or pulsed ultrasonic signal is sent out during the shaping cycles in such a way that the continuous or pulsed ultrasonic signal passes through the melt, - the continuous or pulsed ultrasonic signal that has passed through the melt is measured while generating a large number of measurement signals, a parameter representative of a state variable of the melt is determined from the measurement signals from at least two of the shaping cycles, and the parameters representative of the state variable for the at least two shaping cycles are compared with one another. 12. The method according to claim 11, characterized in that the parameters representative of the state variable are compared with one another with regard to a tolerance range and preferably molded parts that were produced in those molding cycles whose parameters representative of the state variable are outside the tolerance range are declared as rejects will. 13. The method according to claim 11 or 12, wherein a method according to one of claims 3 to 10 is carried out in each of the molding cycles. 14. The method according to any one of the preceding claims, characterized in that due to the detected foreign objects and / or incompletely melted or plasticized partial volumes and / or deviations from state variables - a message is issued to an operator and / or - an automatic adjustment of a machine setting is carried out will. 15. Use of a plasticizing unit for a molding machine and / or a molding machine in a method according to one of the preceding claims. In addition 6 sheets of drawings