WO2005078398A1

WO2005078398A1 - Method and device for the capacitive determination of a filling level

Info

Publication number: WO2005078398A1
Application number: PCT/EP2005/050640
Authority: WO
Inventors: Igor Getman
Original assignee: Endress+Hauser Gmbh+Co. Kg
Priority date: 2004-02-18
Filing date: 2005-02-14
Publication date: 2005-08-25
Also published as: DE102004008125A1

Abstract

The invention relates to a method for capacitively determining the filling level of a medium (1) in a container (2). According to said method, at least one level probe (3) and at least one control/evaluation unit (4) are provided, the latter triggering the level probe (3) by means of an electrical triggering signal and evaluating an electrical response signal of the level probe (3). The inventive method is characterized in that a modification of at least one process condition is determined, an evaluation algorithm is used for determining the filling level in accordance with the process condition resulting from the modification, said evaluation algorithm being selected such that the process condition resulting from the modification has minimal effects on determining the filling level, and the filling level of the medium (1) is determined from the response signal and/or a signal that is proportionate thereto via the evaluation algorithm. The invention further relates to a corresponding device.

Description

Description Method and device for capacitive level determination

The invention relates to a method and a device for the capacitive determination of the level of a medium in a container, at least one level probe being provided, and at least one control / evaluation unit being provided which controls the level probe with an electrical control signal , and which evaluates an electrical response signal from the level probe.

Capacitive measuring devices for level measurement have been known for many years. One probe protruding into the container and the container wall or two probes protruding into the container form a capacitor. Its capacitance C is at least dependent on the fill level and the dielectric constant of the medium to be measured. The fill level can thus be determined from the capacitance C. However, there are some difficulties.

One possibility for measuring this capacitance C is the so-called apparent current measurement. Here, e.g. by means of a conventional rectifier circuit, the amount of the alternating current is measured, which flows at a certain frequency and voltage through the capacitor of the capacitance C to be determined, which is formed from the probe, medium and container wall. However, the current I is not only dependent on the capacitance C, but also on the conductivity s of the medium to be measured. Since the conductivity s of bulk goods in particular depends on various factors such as Depending on temperature or humidity, this results in inaccuracies.

[004] One method of suppressing the influence of this parallel conductivity is the measurement at relatively low frequencies. The portion of the apparent current I flowing through the capacitance C is proportional to the frequency, whereas the portion of the conductivity sverursadite remains constant. Thus, at high frequencies, the capacitive component almost always predominates. Experience has shown that measuring at high frequencies (> 100 kHz) leads to difficulties with long probes with large parasitic inductances.

Another method for measuring the capacitance C is not to measure the current I but the reactive current at a phase shift angle of 90 ° between current and voltage, which corresponds to a pure capacitance measurement. This can be done, for example, with the help of a synchronous rectifier Realize circuit (see patent DE 42 44 739 C2). This method is associated with disadvantages for individual media. With media with a low dielectric constant and high conductivity, which can be easily measured with a current inflow measurement, difficulties arise due to the practically vanishing reactive current. In addition, experience has shown that such conventional synchronous rectifier circuits are sensitive to electromagnetic interference.

Problems with the measurements continue to arise e.g. due to tolerances of the components used and e.g. by approach that can occur on the measuring probe through the medium to be measured. This approach influences the measurement signal and thus also the measurement value, i.e. the level to be determined.

In general, there is also the problem that the capacitance C of the capacitor consisting of the probe and the container wall / second probe depends on many process variables or process conditions - since the fill level is determined in the present invention, process conditions are understood to mean all process variables but the fill level is: fill level, dielectric constant and conductivity of the medium, geometry of the probe, position of the probe relative to the container wall or to the second probe, temperature and pressure in the container etc. If one of these process conditions changes, the capacitance value C can also change. If it is not ensured in the process that only the fill level changes, changes in the capacitance C are no longer clearly associated with changes in the fill level. This also applies to level measurements for which - under different process conditions - reference values for the calibration have been determined.

The invention has for its object to determine the level of a medium as accurately and reliably as possible. A method and a device are required for this.

[009] According to the invention, the object is achieved in that a change in at least one process condition is determined, in that, depending on the process condition resulting from the change, an evaluation algorithm is used for the determination of the fill level such that the change resulting from the change The resulting process condition has minimal effects on the determination of the fill level, and that the fill level of the medium is determined from the response signal and / or a signal proportional to it via the evaluation algorithm. In the simplest case, the evaluation algorithm is a corresponding formula with which the fill level can be calculated from the sizes of the response signal (eg phase, amplitude, amount). The level is therefore in each case elaborated with such a formula / algorithm that optimally adapts to the prevailing process conditions - conductivity of the medium, possible approach to the level probe, temperature, geometry of the probe and also position of the probe in relation to the container, etc. - in such a way is that the process conditions have no impact, especially that they do not cause non-linearities. The change in a process condition can be determined, for example, by the user, who then transmits corresponding data to the measuring device. The determination is thus made by evaluating the value entered by the user or by an additional measuring point or a description of the process conditions associated therewith. The ideal evaluation algorithm / formula for the given situation lets you select from a list of algorithms stored in the measuring device, for example, or an individual adaptation to the prevailing process conditions would also be possible via optimization. For this purpose, for example, a data line to a remote evaluation unit can be provided, which may have a larger rediner capacity. However, it can also be provided that a limited selection of predefined evaluation algorithms / formulas is stored appropriately, from which a suitable algorithm is selected by the user or by the measuring device itself. In the event that the monitored process condition is the approach, a mode A could be provided for no approach up to a maximum of x mm and mode B for an approach between x mm and a maximum of y mm. A too large approach - as explained below - cannot be compensated for by the type of evaluation, so that it is imperative to eliminate the approach in order to be able to measure correctly again. If all process conditions were constant, one would be optimally adapted

Formula suffice. In order to avoid non-linearities, it may make sense to limit the area of application of the measuring device, for example with regard to the media with regard to their conductance. The problem, however, is that most process conditions are not constant, but change. Therefore, the idea of the invention is that the evaluation algorithm / formula is adapted to the changed conditions in such a way that the changed process condition has minimal, ie in the best case no effects on the determination of the fill level. The process condition that has the greatest changes or the greatest influence on the level determination is preferably selected. It is furthermore preferably the process condition that sidi does not restrict, such as the input Direct application to certain media or temperature range. An example of this is the approach that results from the process. With the approach, however, a warning message could also be issued to the user if the approach is too large, so that cleaning is absolutely necessary.

An advantageous embodiment of the method according to the invention provides that the response signal and / or a signal proportional to it is digitized in the control / evaluation unit. A digitization of the response signal for the capacitive level determination is described, for example, in the applicant's patent application to the German Patent and Trademark Office under the file number 103 22 279.0. Such digitization allows the response signal to be optimally accessible to the evaluation. It is not the response signal itself, which is usually a current signal, that is digitized, but rather the voltage signal proportional to it, which can be generated from the response signal via a resistor.

An advantageous embodiment of the method according to the invention includes that the amplitude and / or the phase and / or the amount of the response signal is / are used to determine the fill level. From the response signal - usually a current signal - or a signal proportional to it - from the current signal a voltage signal is usually generated via a resistor - so up to three variables are available, from which the change in a process variable or from which the Level can be calculated. In the prior art, usually only the amount of the response signal is evaluated, which means a loss of information. In this embodiment, there are therefore up to two additional variables, by means of which further and more precise calculations are possible. However, this presupposes that the response signal audi is hardly evaluated with regard to the phase, which is easy to implement through digitization. The evaluation of the phase of the response signal or a signal proportional to it has the advantage that the phase usually gives a large dynamic range, which often exceeds the dynamic range of the amplitude. However, this depends on the properties of the medium whose level is to be measured and / or monitored.

An advantageous embodiment of the method according to the invention includes that the change in a process condition is determined by evaluating the response signal at a known fill level value of the medium. Is the fill level of the medium known - for example via a second fill level measuring device or via the input of value - so conclusions can be drawn. Especially when there are two response signals that were obtained at different times with the same level value, a statement about the change in process conditions can be obtained from the comparison of the signals. If an approach has occurred, for example, this can be reflected in a different phase. The idea is therefore that with the same fill level, different response signals or a response signal with predetermined target values are compared. If there are differences, this must be due to changes in the process conditions and the evaluation is preferably carried out from this point in time using an adapted algorithm or an adapted evaluation formula.

[014] An advantageous embodiment of the method according to the invention provides that a change in an approach to the level probe is determined. Depending on the nature of the medium, it can adhere to the probe. Approach occurs when the medium hangs on the probe after the level has dropped and e.g. dries. Such an approach then acts like a metal sheath that is pulled over the probe. As a result, the probe is virtually shielded and it is hardly possible that the level can no longer cause a change in capacity. Therefore approach is a very important process condition, which is subject to changes especially during the change in the fill level, depending on the temperature or also the time over which the approach exists on the probe.

An embodiment of the method according to the invention provides that a change in the approach is determined from the phase of the response signal or the signal proportional to it. Approach results from the fact that the medium remains attached to the probe. This results in a kind of shielding around the probe. Now, not all processes allow the probe to be cleaned regularly, i.e. to free it from build-up. Therefore, the approach should be considered in the level determination. A first approximation shows that the approach leads to a kind of phase shift. It is therefore obvious that a formula is used to calculate the fill level, which among other things subtracts a phase value from the phase of the response signal. The phase shift is thus virtually canceled. However, this phase value to be subtracted depends on the extent and also on the other nature of the approach, e.g. also whether it is an approach from above or from below, i.e. the medium on the rod is up.

An embodiment of the method according to the invention includes that the filling Stand probe is controlled with an AC signal as an electrical control signal, and that the frequency of the control signal is changed depending on the process condition resulting from the change. Depending on the frequency of the control signal, it is possible that the approach in particular has more or less influence on the determination of the fill level. Thus, the idea of this embodiment is that the frequency of the control signal is changed, preferably increased, as a function of determining an approach. However, this configuration is limited by the current available and by the problem that the fill level probe can function depending on its length from the antenna. With this configuration, the evaluation algorithm is based on the fact that a different excitation frequency compensates for negative consequences, for example of the approach.

An embodiment of the method according to the invention includes that in the event that the process condition resulting from the change lies outside of a predetermined range, an alarm is issued. It is e.g. it is possible that an approach becomes so thick that a reliable measurement is no longer possible because, for example, the response signal is too weak. It therefore makes more sense for the user to be informed that cleaning of the probe is absolutely necessary. This applies not only to the approach, but also to changes in the conductivity of the medium, the temperature, the pressure, etc.

The invention also provides a device for capacitive determination of the level of a medium in a container, at least one level probe being provided, and at least one control / evaluation unit being provided which controls the level probe with an electrical control signal, and which evaluates the electrical response signal of the level probe.

[019] The invention achieves the object with respect to the device in that the control / evaluation unit is assigned at least one memory unit in which evaluation algorithms for determining the fill level are stored. The control / evaluation unit is therefore assigned a feeder unit - this storage unit can be part of the actual device or connected to the device via a data communication type (fieldbus, Ethernet, etc.) - in which several evaluation algorithms or simple evaluation formulas are stored or can be generated appropriately based on other stored data. This means that several algorithms / formulas are available for determining the level, whereby the optimal algorithm or formula is used to change a process condition. An advantageous embodiment of the device according to the invention provides that the control / evaluation unit is designed such that it determines the change in a process condition and that it uses such an evaluation algorithm for determining the fill level in accordance with the process condition resulting from the change that the process condition resulting from the change has minimal effects on the level determination. The control / evaluation unit thus determines the change in a process condition and then determines the fill level with an algorithm. In the case of weldiem, the changed and thus new process conditions have as little effect as possible, for example nonlinearities should be avoided as far as possible. The change in a process condition can hardly be determined, for example, via an additional sensor which monitors a process condition. It is more elegant to determine the change in a process condition directly from the response signal. Approach, for example, affects the phase with regard to extreme values and also with regard to the course of the phase. Thus, for example, the approach can be drawn from the phase.

An advantageous embodiment of the device according to the invention provides that the control / evaluation unit is assigned at least one analog / digital converter which digitizes at least the response signal and / or a signal proportional to it. This allows the response signal - usually a current signal - or a signal proportional to it - usually a voltage signal - to be digitized and thus optimally evaluated, which also includes that the response signal not only evaluates the amplitude but also the phase. The control / evaluation unit is preferably a microprocessor with a corresponding signal input to the analog / digital converter.

One embodiment of the device according to the invention provides that at least one additional fill level measuring device is provided for determining the fill level of the medium, the measured value of which is used by the device for determining the change in a process condition. The change in a process variable can be determined using this additional level measuring device, e.g. different response signals with the same level value can be compared with one another or with corresponding target values. Another embodiment is that the user transmits the fill level value of the device via an input unit and thus contributes to determining the change in a process condition.

An embodiment of the device according to the invention includes that the control / evaluation unit is designed in such a way that it comes from the phase of the response signal or a signal proportional to this determines a change in the approach to the level probe. This is the configuration mentioned above that the change in the process condition itself is determined from the response signal. The response signal is therefore preferably checked for changes that do not result from the change in the fill level, but rather in the process condition. One possibility is to check the minimum or maximum values of phase or amplitude. Another possibility is to evaluate the course of the variables over time or as a function of the fill level. The phase is preferably examined in order to infer the approach from this.

The invention is explained in more detail with reference to the following drawings.

[025] It shows:

1: shows a schematic structure of a capacitive

[027] level measurement,

[028] FIG. 2: a flow chart to clarify the method,

3: a diagram to clarify the dependency

[030] the admittance of the response signal from the conductivity and the

DK value of the medium,

4 shows a diagram to illustrate the effect of the

Approach to Admittance, and

5: a diagram to illustrate the errors of two

[035] different evaluation formulas.

In Fig. 1, the level probe 3, the wall of the container 2 and the medium 3 form a capacitor, the capacitance C, inter alia, is due to the level of the medium 1. Other variables that influence the capacitance C include the conductance and the dielectric constant (DK value) of the medium 1, the temperature and the pressure. There are also dependencies on the geometry of the probe 3, the container 2 and the positioning of the probe 3 and the container 2 to one another. Furthermore, depending on the type of medium 1, formation 5 on probe 3 or the conversion of container 2 may occur. Not only does the level of the medium 1 change the capacitance C, but other process conditions also have an influence. Conversely, since this makes the determination of the fill level difficult, uncertain or even impossible, the device of the invention is further developed. In the control / evaluation unit 4 - this can be a component of the measuring device directly, but can also be further away from it and connected to the probe 3 or an electronic unit connected in between, for example via a data line first the response signal of the probe 3 or a signal proportional to it - a voltage signal can be obtained from the current signal via a resistor (not shown graphically here) - digitized by an analog / digital converter 11. For this purpose, a microprocessor is provided in the control / evaluation unit 4, for example. From this digitized signal, it can be determined, for example, whether attachment 5 has formed on probe 3 or whether the extension of attachment 5 has changed. If there are changes - other process conditions can be overridden; appropriate sensors may be required for this, if the response signal does not show a behavior that is clear from the process condition - the fill level is determined from the response signal using a formula adapted to the new process conditions. Corresponding formulas and evaluation algorithms are stored in the memory unit 10. There, audi data can be stored, from which an optimally adapted formula or an optimal algorithm results. The optimization consists in the fact that the process conditions resulting from the change have as little influence as possible on the determination of the fill level, that is to say in particular they do not produce any incorrect values. However, the storage unit 10 does not have to be a component of the preparatory device or the control / evaluation unit 4, as shown here. The idea of the invention is therefore to adapt the evaluation algorithm / the formula for determining the fill level and thus the determination itself to the prevailing process conditions. However, this requires recognizing that the process conditions have changed. Since such process conditions can also occur which lie outside certain limits, ie which lie outside the range in which meaningful and reliable measurements are possible, a display unit 15 is also provided in the embodiment of the device for implementing the method shown in the figure, It is signaled via weldie, for example, that the approach is so large that the level probe 3 must be cleaned. The limits of the process conditions must be specified accordingly and also depend on which process conditions due to the electronics and the evaluation algorithms and formulas still allow an evaluation of the level signal. The fill level measuring device 20 is, for example, a fill level limit switch, which is manufactured and sold by the applicant under the name "Liquiphant". This measuring device 20 is positioned, for example, relative to the probe 3 in such a way that it detects when the fill level of the medium 1 is such that the Probe 3 is no longer covered by medium 1, so if the probe should be free of medium, Durdi Approach 5 is probe 3 from Medium 1 is surrounded by medium 1 even when it is actually free / uncovered. Since the fill level measuring device 20 is connected to the control / evaluation unit 4, the measured values / signals of the secondary measuring device 20 can be used to determine the change in a process variable - here preferably the approach. By comparing response signals that result when the fill level is below the lower end of the probe, conclusions can be drawn about approach 5, for example by evaluating the phase. Instead of the additional measuring device 20, it is also possible for a fixed level value to be transmitted through an interaction with a user of the control / evaluation unit 4. What is generally required is a quantity which enables the comparison of response signals and which thus compensates for the uncertainty or the error which results from the change in the process condition.

Figure 2 shows that the critical point of the invention is to determine whether a process condition has changed. This can be the approach, the conductance, the geometry, the position of the probe relative to the container wall - changes are e.g. possible by operating an agitator -, the temperature, the pressure or other process conditions. In most cases, the focus will be on a specific process condition that is subject to the greatest changes or that shows the greatest changes in determining the fill level. For example, this is an approach e.g. of the medium on the probe. Jedodi audi can monitor several process conditions. However, it is then necessary to weigh the benefits against the effort.

The method is as follows: if there is no change, the fill level is further determined. If there is a change in the fill level, for example the approach has become larger, an algorithm or a formula is used for the further determination of the fill level, which is adapted to the new process condition. The process condition resulting from the change in the process condition is therefore the new or current process condition, the change in one condition also being able to influence other process conditions. For example, the change in temperature and the influence of the mixture on the response signal cause the mixture to dry out. The fill level is then also determined using the new algorithm. The change in a process condition can be determined, for example, from an extreme value in the response signal, or, for example, from the examination of the course of a series of measurements. The values in FIGS. 3 to 6 are elaborated and simulated on the basis of measurement results. The dependence of the admittance of the response signal on the conductivity s (X-axis) and the DK value (the individual curves) of the medium is shown in FIG. 3. The conductivity of the medium is the reciprocal specific resistance of the medium: s = 1 / r. The DK value (dielectric constant or relative permittivity number) of the medium e is used to calculate the capacitance C which results from the probe, the second probe or the container wall and the medium: C ~ e. The following DK values are shown (from the bottom to the top): DK = 2, 10, r 20, 50 and 80. The admittance A or the admittance is the amplitude of the response signal. It can be seen from the graphic that the admittance depends on the material properties such as conductivity and DK value. The dependence on the frequency of the excitation signal is also shown: this is 33 kHz for the solid lines and 1 MHz for the dashed lines. The frequency can therefore be used to achieve a shift that corresponds to a change in the conductivity of the medium in certain areas. This therefore shows that the design of the measurement or the measurement parameters makes it possible to react to the given conditions. It is not always possible for Jedodi to set the frequency arbitrarily, since this also depends on the available energy and because care must be taken to ensure that the probe is not converted into a radiating antenna, so that it does not reach the resonant frequency of the probe becomes. However, the dependence of the admittance on the properties of the medium also means that the range of application of the capacitive level determination is not directly possible with all media. It may therefore be necessary to design the capacitive measuring device to match the specific medium.

[041] The process condition approach exists when the medium adheres to the level probe. The approach then has the effect as if a metallic tube had been put over the probe. This leads to the fact that the medium can no longer be measured correctly outside of the approach. The probe is thus virtually shielded by the approach. It has been shown that in addition to the purely geometrical dimensioning of the approach, the temperature of the medium in the approach, but also the degree of drying of the approach, have a strong impact on the response signal from the probe.

4 shows the effects of the approach. The admittance is shown depending on the conductivity of the medium. The DK value of the medium is 59 and the fill level height 60% (0% corresponds to a probe not covered by the medium and 100% to a fully covered probe; the absolute values therefore depend on the length of the probe and the position of the end of the probe relative to the bottom of the container). Curve No. 1 shows the course of the admittance as a function of the conductivity at a frequency of 250 kHz without starting. The approach to the probe results in curve no. 2. In curve 2, from a conductivity of 1000 mS / cm, there is a dramatic increase in the admittance through the approach. The approach therefore results in an excessively high admittance and, conversely, the admittance leads to an excessively high fill level, which is not given. In this example, the fill level is over-over-eaten by the probe. Dashed curve # 3 shows the effect of using a higher frequency of 1 MHz at approach. Due to the higher excitation frequency, the over-etching of the fill level takes place only with higher conductivities. The change in frequency thus provides a larger range for media that can still be measured reliably despite the approach. Curve No. 4 uses the formula F2 (more on this in a moment), although the excitation frequency is still 250 kHz. It can thus be seen that the evaluation formula also results in a shift that comes close to the use of a higher frequency. Because of the type of evaluation, the possibilities of measurement can be significantly expanded. 5 shows the relative measured fill level (m for measured; H for the height and L for the length of the probe; thus H / Lm) to the relative actual fill level (H / L). The dashed line thus represents the optimal course of the measurements. The conductivity should be 5000 mS / cm and the approach thickness is 1 mm with a rod length L of 500 mm. The approach here is calculated so that it covers the entire probe. The length of the approach is 500 mm. It should also be noted that the effect of varying the thickness of the attachment depends on how long the probe is. The excitation frequency is 250 kHz. Curve 1 was with the evaluation formula

Fl calculated: Fl = A sin f + A cos f cos (f - f). A is the admittance and f is the o

Phase of the measured response signal, f is a constant phase shift that results from the optimization of the evaluation curve. Curve 2 was elaborated with the formula F2: F2 = A (kl sin f - k2 cos f). This evaluation formula therefore uses two constants kl and k2, the optimization of which results in a possibly good adaptation of the curve to the target values. As can be seen, Fl generally leads to an overestimation of the fill level. The function F2 usually delivers minor errors, has jedodi too This also leads to an underestimation of the fill level. It is therefore obvious that depending on the selection of the evaluation formulas or the evaluation algorithm in general, the measured values can be better and more accurately evaluated and that it is possible to compensate for adverse process conditions such as the approach by the type of evaluation.

Claims

Claims [001] Method for capacitively determining the fill level of a medium (1) in a container (2), at least one fill level probe (3) being provided, and at least one control / evaluation unit (4) being provided, which the fill level probe ( 3) controlled with an electrical control signal and which evaluates an electrical response signal from the fill level probe (3), characterized in that a change in at least one process condition is determined, depending on the process condition resulting from the change, for determining the fill level Such an evaluation algorithm is used that the process condition resulting from the change has minimal effects on the determination of the fill level, and that the fill level of the medium (1) is determined from the response signal and / or a signal proportional to it via the evaluation algorithm.

[002] The method according to claim 1, characterized in that the response signal and / or a signal proportional thereto is digitized in the control / evaluation unit (4).

[003] Method according to claim 1 or 2, characterized in that the amplitude and / or the phase and / or the amount of the response signal is / are used for determining the fill level.

[004] Method according to claim 1 or 4, characterized in that the change in a process condition is determined by evaluating the response signal at a known fill level value of the medium (1).

[005] The method according to claim 1, characterized in that a change in an extension (5) on the level probe (3) is determined.

Method according to claim 5, characterized in that a change in the approach (5) is determined from the phase of the response signal or the signal proportional thereto.

The method of claim 1, 5 or 6, characterized in that the level probe (3) is controlled with an alternating current signal as an electrical control signal, and that the frequency of the control signal is changed depending on the process condition resulting from the change.

[008] The method of claim 1, 5 or 7, characterized in that in the In the event that the process condition resulting from the change lies outside a predetermined range, an alarm is issued.

Device for the capacitive determination of the fill level of a medium (1) in a container (2), at least one fill level probe (3) being provided, and at least one control / evaluation unit (4) providing the fill level probe (3 ) with an electrical control signal, and which evaluates an electrical response signal from the fill level probe (3), characterized in that the control / evaluation unit (4) is assigned at least one feed unit (10) in which evaluation algorithms for determining the fill level of the Medium (1) are stored.

Vorriditung according to claim 9, characterized in that the control / evaluation unit (4) is designed such that it determines the change in a process condition and that it a sold en evaluation algorithm for determining the level according to the process condition resulting from the change uses that the process condition resulting from the change has minimal effects on the determination of the fill level.

Device according to claim 9, characterized in that the control / evaluation unit (4) is assigned at least one analog / digital converter (11) which digitizes at least the response signal and / or a signal proportional thereto.

Vorriditung nad claim 9, characterized in that at least one additional level measuring device (20) is provided for determining the level of the medium (1), the measured value of which is used by the device for determining the change in a process condition.