METHOD FOR OPERATING A GROUP OF PRESSURE SENSORS

Description

[0001] The present invention relates to a method for operating a group of at least two pressure sensors.

[0002] Various measurement principles for measuring pressure are known in the state of the art. These measurement principles can be divided into two groups: direct pressure measurement principles, which ultimately determine a force per area, and indirect pressure measurement principles, which exploit the dependence of another physical quantity on pressure, for example, the pressure dependence of the thermal conductivity of a gas. Indirect pressure measurement principles exhibit a greater or lesser dependence on the type of gas.

[0003] Examples of pressure sensors that use a direct, gas-independent pressure measurement principle include piezo-diaphragm manometers, capacitance-diaphragm manometers, or optical diaphragm manometers. Examples of pressure sensors that use an indirect, gas-dependent pressure measurement principle include Pirani sensors, cold-cathode ionization vacuum gauges (e.g., inverted magnetron), or hot-cathode ionization vacuum gauges (e.g., Bayard-Alpert).

[0004] Pressure sensors are usually designed for a specific pressure measurement range due to their measuring principle. The measuring range can be extended by combining several pressure sensors into a pressure sensor group. However, in the case of a combination of pressure sensors that all use an indirect pressure measurement principle, the measurement uncertainty is increased by the gas type dependence, as the gas type dependence can vary between the pressure sensors involved. Furthermore, it is often not known exactly which gas or gas mixture is present.

[0005] The following documents show the state of the art.

[0006] EP 1 394 523 A1 relates to a method for operating a total pressure transmitter consisting of at least two sensor elements. In a higher pressure range of approximately 10 -3 to 1000 mbar, the pressure is measured with a first sensor element. A second sensor element, usually a hot-cathode sensor, remains inoperative during this time. This protects it and extends its service life. To calibrate the lower support point of the first sensor element, it is sufficient to switch on the hot-cathode system at longer intervals.

[0007] DE 198 60 500 A1 relates to a device for pressure measurement comprising a first pressure sensor for detecting a pressure in a first pressure range, a second pressure sensor for detecting a pressure in a second pressure range, wherein the first and the second pressure range overlap, and a calibration device for calibrating the second pressure sensor by means of the first pressure sensor by correcting the output of the second pressure sensor based on the outputs of the first and the second pressure sensor within the overlap region of the first and the second pressure range, such that the calibrated output of the second pressure sensor in the overlap region is substantially equal to the output of the first pressure sensor.This allows calibration of the second pressure sensor in a simple yet accurate manner, which is particularly advantageous when pressure measurements have to be carried out in the vacuum range, where the first pressure sensor is an absolute pressure sensor, and the second pressure sensor is, for example, a heat conduction pressure sensor or an ionization pressure sensor, etc., which depends on the type of gas.

[0008] EP 0 379 841 A2 presents a method for measuring pressure, preferably vacuum. The measurement signals from a heat conduction manometer and measurement signals from a gas friction manometer are preferably processed by multiplication to form a common measurement signal. They are linearly superimposed or added. Preferably, a Pirani manometer is used as the heat conduction manometer, and a tuning fork quartz manometer is used as the gas friction manometer.

[0009] The object of the present invention was to mitigate the difficulties arising from the gas dependence of pressure sensors. The object of the present invention was, in particular, to provide a method for operating a group of pressure sensors that reduces measurement inaccuracies arising from the gas type dependence of pressure sensors.

[0010] This object is achieved by a method according to claim 1.

[0011] The method according to the invention is a method for operating a group of pressure sensors. The pressure sensors belonging to the group of pressure sensors are arranged such that they can measure the pressure in a common measuring volume. The group of pressure sensors comprises at least a first pressure sensor with a first pressure measuring range and a second pressure sensor with a second pressure measuring range, wherein the first and second pressure measuring ranges overlap in an overlap pressure measuring range. The first pressure sensor is based on a first indirect pressure measuring principle and is configured to output a first measuring signal calibrated to a reference gas, e.g., nitrogen. The second pressure sensor is based on a second indirect pressure measuring principle and is configured to output a second measuring signal calibrated to the same reference gas.

[0012] The procedure includes the following steps: a) Providing gas-type-specific first calibration data for the first measurement signal and gas-type-specific second calibration data for the second measurement signal, wherein the first and second calibration data describe a dependence of the first and second measurement signals, respectively, on the effective pressure and on a gas type in the common measurement volume for a list of gas types comprising at least one first gas type that is different from the reference gas; b) substantially simultaneously detecting a first measured value of the first measurement signal and detecting a second measured value of the second measurement signal; c) determining a resultant gas type as the gas type in the list of gas types that, taking into account the first and second calibration data, best matches the combination of the detected first measured value and the detected second measured value.

[0013] The first and second calibration data can be defined, for example, as a combination of a mathematical function and at least one parameter of this function. Alternatively, the calibration data can be defined as table values ​​(in a look-up table). A calibration curve can be read by interpolating the table values.

[0014] The inventors have recognized that, with the present invention, information about the gas type present in the common measurement volume can be obtained in a simple manner. This information can, as discussed further below, be used to improve the accuracy of pressure measurements based on the measurement signals from the pressure sensors. However, this information about the gas type is already valuable in itself. When operated according to the method of the invention, the group of pressure sensors can, in a surprisingly simple manner, at least partially assume the function of a residual gas analyzer.

[0015] The list of gas types can be a list of pure chemical substances, such as nitrogen (N 2 ), oxygen (O 2 ), helium (He), argon (Ar), xenon (Xe), carbon dioxide (CO 2 ), water vapor (H 2 O), etc. The list of gas types can also include a list of gas mixtures. For example, different mixing ratios of the same gases can have separate entries in the list, each with its own calibration data. Such a list can, for example, include entries for 100% N 2 , 90% N 2 + 10% Ar, 80% N 2 + 20% Ar, etc. Depending on the sensitivity of the calibration data to the gas concentrations, 5%, 2%, 1%, etc. steps can also be provided in order to provide calibration data that describe the gas type dependency with sufficient accuracy. The list of gas types can summarize groups of different chemical substances in one list entry.For example, in the case of a Pirani sensor, the calibration data for air, nitrogen, and oxygen can be treated together. For example, the list of gas types can also be defined by a list of molecular weights. Chemical substances with the same molecular weight are recorded in a single list entry.

[0016] For the purposes of the present invention, the term "list of gas types" should also be understood as a generalized form of a list in which the list entries are determined not by a finite number of list indices, but by a continuously running parameter. This parameter can, for example, be a mixing ratio of gases. As a further example, this parameter can be an average molecular weight. In this case, first and second calibration data are defined for each value of the parameter as a function of this continuous parameter using a mathematical formula or an interpolation of table values. This generalized form of a list of gas types plays a role in particular in variants of the inventive method based on balancing methods ("best fit"), which will be mentioned later.

[0017] The calibration data can, for example, be measured data. Alternatively, the calibration data can also be derived from theoretical properties of the pressure sensor. The calibration data can, for example, be based on a computer simulation, particularly taking into account the geometric shapes and dimensions of the respective pressure sensor. Embodiments of the method emerge from the features of the dependent claims 2 to 11.

[0018] A variant of the method comprises the additional step d) determining a result pressure as a function of the recorded first measured value and the first calibration data for the result gas type and / or as a function of the recorded second measured value and the second calibration data for the result gas type.

[0019] The inventors have recognized that, based on the previously determined information about the gas type present, high measurement accuracy can be achieved across the entire pressure measurement range, which is the combined pressure measurement range of at least two pressure sensors in the group. The uncertainty of the determined pressure, which arises from the lack of knowledge about the gas type present in the measurement volume, is reduced in this variant of the method or, ideally, even eliminated completely. In this variant, a gas-type-independent result pressure is also determined.

[0020] In a variant of the method, the first and second pressure sensors are vacuum pressure sensors.

[0021] Vacuum pressure sensors benefit particularly from the method according to the invention, because the lower the pressure to be measured, the more likely it is that an indirect pressure measurement method must be used, which shows a gas type dependency.

[0022] In a variant of the process, the first pressure sensor is a Pirani sensor.

[0023] Pirani sensors exhibit a typical gas type dependency, which, for example, with a sensor calibrated to nitrogen as the reference gas, may result in a measured value for water vapor that is a factor of 2 too high, and for the noble gas xenon, a measured value that is a factor of 3 too low. Without the method according to the invention, this is a source of significant measurement uncertainty. In conjunction with the method according to the invention, this gas type dependency becomes a useful source of information.

[0024] In one variant of the process, the second pressure sensor is a hot-cathode ionization vacuum gauge. In particular, a Bayard-Alpert type hot-cathode ionization vacuum gauge is suitable. Using this measurement principle, a sensor calibrated for nitrogen will indicate a pressure 2.4 times too low for hydrogen, while indicating a pressure 2.5 times too high for xenon.

[0025] This variant can be combined, in particular, with a variant that uses a Pirani sensor as the first pressure sensor. The completely different pressure measurement principles of the two sensors complement each other perfectly, as their gas-type dependencies differ significantly. A change in the gas type in the shared measurement volume manifests itself in a pronounced deviation between the two pressures output by, for example, a Pirani sensor and a hot-cathode sensor calibrated for N2. In the case of xenon, for example, a Pirani sensor calibrated for N2 displays a value that is 3 times too low, while the hot-cathode sensor calibrated for N2 displays a value that is 2.5 times too high.

[0026] In one variant of the method, the second pressure sensor is a cold cathode ionization vacuum gauge. The cold cathode ionization vacuum gauge can, in particular, be an inverted magnetron.

[0027] This variant also works well together with a Pirani sensor as the first pressure sensor.

[0028] In a variant of the method, the first and second gas-specific calibration data are each defined by a first or second factor, respectively, by which the first measurement signal or the second measurement signal is to be multiplied to obtain the effective pressure.

[0029] For example, for the measurement signal p 1 of the first sensor and the i'th gas G i, the formula for the effective pressure p eff = C 1 G i p 1 and for the measuring signal p 2 second sensor p eff = C 2 G i p 2 .

[0030] A table with specific factors for a list of gas types and for the Pirani and cold cathode pressure sensor types is given below.

[0031] In a variant of the method, a list of quotients is formed by forming a quotient from the first factor for the respective gas type and the second factor for the respective gas type for each gas type from the list, wherein a recorded quotient is formed as a quotient of the recorded first measured value and the recorded second measured value, and wherein in step c) "determining a result gas type" it is determined which of the quotients from the list of quotients the recorded quotient is closest to.

[0032] That is, a list of quotients Q[G i ] is defined as Q[G i ] = C 1 [G i ] / C 2 [G i ] or as its inverse, for each i from the list of gases G i .

[0033] The recorded quotient is calculated from the recorded measured values ​​p 1 (of the first pressure sensor) and p 2 (of the second pressure sensor) as Q = p 1 / p 2 .

[0034] Q* is determined as the quotient from the list Q[G i ] which is closest to Q. The corresponding gas type is then the result gas type G*.

[0035] In a variant of the method, in step c) "Determining a result gas type," the expected value for the second measurement signal is determined for each gas from the list of gases, based on the first measured value and the first and second gas-specific calibration data, if this gas were present in the common measurement volume. The smallest deviation of this value from the second measured value is used as the criterion for determining the result gas type.

[0036] With this variant, for example, a selection can be made from a finite list of gas types. This variant is also suitable for the generalized form of a list of gas types mentioned above. The criterion of the smallest deviation between the value predicted from the calibration data for the second measurement signal and the recorded second measured value results in the best choice for the continuous parameter in the list of gas types, for example, the value of a mixing ratio or an average molecular weight. This choice of the continuous parameter defines the resulting gas type.

[0037] In a variant of the method, a plurality of pairs each comprising a first measured value from the first pressure sensor and a second measured value from the second pressure sensor are recorded when the pressure in the common measuring volume changes, wherein when determining the resultant gas type, the gas type is selected which best matches the combination of the recorded plurality of pairs.

[0038] This variant of the method is particularly useful when the calibration curves of different gases differ in their dependence on the effective pressure, e.g., due to different gradients or curvatures, which cannot be observed with a single measurement at a single pressure value. The condition that the pressure in the common measurement volume changes can be actively achieved, for example, by switching on a pump or opening a valve to admit a gas into the common measurement volume. The condition of changing pressure can alternatively also be achieved, however, by continuously observing, for example, a measurement signal from one of the pressure sensors and starting the method as soon as a sufficiently rapid pressure change rate is observed.In this way, measurement data can be collected from a sufficiently large pressure range in which the different courses of gas-dependent calibration curves are reflected in distinguishable measurement results.

[0039] This variant also allows working with a list of gas types, where the list entries are defined by a continuous parameter. This can further increase the accuracy of the determined pressure, as there is no need to "round" to the most suitable gas from a table. Adjusting a continuous parameter can better reflect the actual situation of gas mixtures present in the measurement volume.

[0040] In a variant of the method, the recorded first measured value, the recorded second measured value or the result pressure are used to check whether the pressure present in the common measuring volume lies within the overlap pressure measuring range and the result pressure and / or the result gas type are rejected as invalid if this is not the case.

[0041] If the effective pressure in the shared measurement volume does not correspond to a pressure within the pressure measurement range of both pressure sensors used for the steps of the method, the determined resulting gas type and, if applicable, the resulting pressure will not be meaningful. Whether the aforementioned condition is met may not become apparent until some steps of the method have already been performed. This variant of the method provides a simple way to ensure quality assurance.

[0042] Furthermore, the invention also relates to a device according to claim 12. This is a device for carrying out the method according to the invention.

[0043] The device comprises a group of pressure sensors arranged such that they can measure the pressure in a common measurement volume. The group of pressure sensors comprises at least a first pressure sensor with a first pressure measuring range and a second pressure sensor with a second pressure measuring range, wherein the first and second pressure measuring ranges overlap in an overlap pressure measuring range. The first pressure sensor is based on a first indirect pressure measuring principle, and the second pressure sensor is based on a second indirect pressure measuring principle. The device further comprises means for storing first calibration data and second calibration data.The device also comprises a control unit operatively connected to a first measurement signal output of the first pressure sensor, to a second measurement signal output of the second pressure sensor, and to the means for storing first calibration data and second calibration data for processing the measurement signals of the pressure sensors. The device is further configured to output the resultant gas type. Alternatively, the device is further configured to (preliminarily) determine the resultant gas type and to determine the resultant pressure based on the (previously determined) resultant gas type and to output the resultant pressure. In this alternative, the resultant gas type remains an internal result, and only the resultant pressure is output via an interface. Thus, viewed from the outside, the device behaves like a single pressure sensor, delivering an accurate measurement result for the pressure regardless of the gas type in the measurement volume.

[0044] Further within the scope of the invention is a computer program product according to claim 13. The computer program product comprises instructions which, when the instructions are executed by a control unit of a device according to the invention, cause the control unit to carry out the steps of the method according to the invention or one of the variants of the method.

[0045] In the following, an example is explained using a group of pressure sensors with a Pirani sensor as the first pressure sensor (calibration data in the form of factor C1) and with a hot cathode ionization vacuum gauge (calibration data in the form of factor C2) as the second pressure sensor with concrete calibration data, which in this case are available as factors.

[0046] In the pressure range from 5*10E-2 to 5*10E-4 mbar, it is a good approximation p eff = C 1 G i * p 1 for pressure sensor 1 and p eff = C 2 G i * p 2 for pressure sensor 2, where the corresponding row of the table below corresponds to the gas Gi. This pressure range also lies within the overlap pressure measuring range of the two pressure sensors. The table applies to nitrogen N2 as the reference gas, so the gas-dependent factor for nitrogen is 1. For the sensors selected here, air (air) and oxygen (O2) behave practically identically to the reference gas N2 and are therefore listed in one row of the table. gas C1 C2 Q=C2 / C1 1 / Q=C1 / C2 Hey 0.8 5.9 7.38 0.14 No 1.4 4.1 2.93 0.34 Ar 1.7 0.8 0.47 2.13 Kr 2.4 0.5 0.21 4.80 Xe 3.0 0.4 0.13 7.50 H2 0.5 2.4 4.80 0.21 air, O2, N2 1 1 1.00 1.00 CO2 0.9 0.70 0.78 1.28 H2O 0.5 0.89 1.79 0.56

[0047] If the effective pressure p eff in the common measuring volume of the two pressure sensors is within the calibration range by a gas-dependent factor according to the table, the first pressure sensor delivers the measured value p 1 = p eff / C 1 and the second pressure sensor delivers the measured value p 2 = p eff / C 2 . The quotient of the two measured values ​​is therefore p 1 / p 2 = C 2 * p eff / C 1 * p eff = C 2 / C 1 , regardless of the exact value of p eff . The overlap pressure measurement range, or the validity range of the approximation, extends over two decades in this example.

[0048] If, for example, a value of over 7 is determined for the quotient p 1 / p 2 , then within the list of gases in the table above, helium (He) fits best (table value for Q[He] = 7.38), while the second best-fitting value (Q[H2] = 4.80) is already relatively far away. In step c) of the inventive method, helium would be determined as the resulting gas type in this case. In step d) of the variant of the method, the resulting pressure p* can now be determined using the factors for the gas helium, i.e. the resulting gas type, e.g., using the formula p * = C 1 He * p 1 = 0.8 * p 1 .

[0049] Alternatively, the formula p * = C 2 He * p 2 = 5.9 * p 2 applied, or an average value based on measured values ​​from both pressure sensors can be used, for example as p * = C 1 He * p 1 + C 2 He * p 2 / 2 be determined.

[0050] Embodiments of the present invention are explained in more detail below with reference to figures. Fig. 1schematically shows a device for carrying out the method; Fig. 2 a flow chart of the method according to the invention; Fig. 3 schematically possible relative positions of the first and second pressure measuring range and the resulting overlap pressure measuring range; Fig. 4 shows in a double-logarithmic representation the dependence of a pressure determined by means of a Pirani sensor on the type of gas; Fig. 5 shows schematically first and second gas-dependent calibration curves.

[0051] In Figure 1An exemplary device 10 for carrying out the method is shown schematically. The device comprises a group 1 of pressure sensors, with at least a first pressure sensor 1' and a second pressure sensor 1", which can measure pressures in a common measuring volume 2. The measuring volume 2 can, in particular, be a partial volume of a vacuum chamber, as schematically indicated by the area outlined in dash-dotted lines. The first pressure sensor 1' is configured to forward a first measurement signal p 1 from a first measurement signal output 3' to a control unit 12. The second pressure sensor 1'' is configured to forward a second measurement signal p 2 from a second measurement signal output 3'' to the control unit 12. The active connections indicated by dashed lines can be implemented, for example, via wires, or they can also be realized, for example, via radio signals (Bluetooth, etc.) or optical signal transmission.Dashed arrows indicate the flow of information between the elements of the device. The device includes means 5 for storing gas-dependent calibration data, which can be transmitted to the control unit. A result gas type G* and a result pressure p* can be output by the control unit.

[0052] The parts of the device shown or the complete device can be installed in a common housing. In particular, the group of pressure sensors and the control unit can be combined in a common housing to form a pressure sensor unit. Additionally, a means for storing the calibration data can optionally be housed in the common housing.

[0053] In Figure 2 A flowchart of the inventive method 100 is shown. The method comprises the steps a) Providing 101 gas type-specific first calibration data K 1 [G i ] for the first measurement signal and gas type-specific second calibration data K 2 [G i ] for the second measurement signal, wherein the first and second calibration data describe a dependency of the first and second measurement signals, respectively, on the effective pressure p eff and on a gas type in the common measurement volume for a list of gas types comprising at least one first gas type G 1 that is different from the reference gas; b) essentially simultaneously detecting 102 a first measured value p 1 of the first measurement signal and detecting a second measured value p 2 of the second measurement signal in the overlap pressure measurement range; c) determining 103 a resultant gas type G* as the gas type in the list of gas types which, taking into account the first and second calibration data, best matches the combination of the detected first measured value p 1 and the detected second measured value p 2 .

[0054] Steps 101, 102, and 103 are performed sequentially, with the necessary calibration data already available before the start of the procedure. At the end of the procedure (END), the resulting gas G* is known.

[0055] The dashed rectangle represents the optional step d), which, if executed additionally, leads to a variant of the procedure that also produces a result print as output. With this additional step, the result print p* is also known at the end of the procedure.

[0056] The additional step d) involves determining 104 a resultant pressure p* as a function of the acquired first measured value p1 and the first calibration data for the resultant gas type and / or as a function of the acquired second measured value p2 and the second calibration data for the resultant gas type. Based on the resultant gas type known from step c), the corresponding set of calibration data is used to translate the measured values ​​of the pressure sensors into the gas-type-independent effective pressure.

[0057] Figure 3 show in Fig. 3.a ) and in Fig. 3.b) schematically shows two possibilities for the relative position of the first 4' and second 4'' pressure measuring range of the first 1' or second 1" pressure sensor of the group of pressure sensors on a pressure axis p. The pressure axis p is to be understood schematically here; it could, for example, be a linear axis or a logarithmic axis. High pressures are drawn further up on the axis than lower pressures. An overlap pressure measuring range 6 exists in which the first 4' and the second 4'' pressure measuring range overlap. The reading of the first and second measuring signals in step a) of the method takes place while the pressure in the common measuring volume lies in this overlap pressure measuring range 6. In Fig. 3.b ) shows the case in which the second pressure measuring range 4'' lies completely within the first pressure measuring range 4', so that the overlap pressure measuring range is identical to the second pressure measuring range 4'.

[0058] Figure 4shows, in a double-logarithmic plot, the dependence of a pressure determined by a Pirani sensor on a specific gas type. The "effective" pressure p eff is plotted horizontally. In the vertical direction, the pressure p (mbar) read from a Pirani sensor is plotted as a function of the effective pressure p eff (mbar) for different gas types, each with a separate curve; see the label for each curve in the upper right area of ​​the graph. The pressure range shown extends on both axes from 10 -3 < mbar to 10 2 < mbar, i.e., over 5 orders of magnitude. The Pirani sensor is calibrated in this case to display the pressure p eff for the gas type air, i.e., the pressure curve for air (Air) is a straight line on the diagonal in the double-logarithmic plot. Each of the curves shown is therefore a gas-type-dependent calibration curve. The first pressure sensor can, for example, be a Pirani sensor, so that the Fig. 4The set of curves shown can represent the gas-specific first calibration data.

[0059] In a pressure range below approximately 1 mbar, the effect of the gas type can be described by a factor between p eff and the pressure p measured with the Pirani sensor, which is shown in the double-logarithmic diagram as an offset of the curves. As the inventors have recognized, the essential information of these calibration curves can be described with sufficient accuracy over approximately two decades by this factor, so that a table with the corresponding factors represents a very memory-efficient form of gas-type-specific initial calibration data.

[0060] Similarly, second calibration data for the second pressure sensor can be provided as a set of curves or as a table of factors.

[0061] Figure 5shows, by way of example and schematically, calibration curves for three gases: for gas G1 (solid line), for gas G2 (dashed with short dashes), and for gas G3 (dashed with long dashes). The left-hand half shows the calibration curves for a first pressure sensor as a function of the first measurement signal p1. The right-hand half shows the calibration curves for a second pressure sensor as a function of the second measurement signal p2. The effective pressure peff corresponding to the respective measurement signal is plotted on the vertical axis, with a specific vertical position in the left-hand diagram corresponding to the same effective pressure as in the right-hand diagram. The calibration curves are to be understood as illustrative examples which clarify the principle on which the invention is based. The diagrams can, for example, be double-logarithmic representations.The gas type dependence of the first sensor differs from that of the second sensor. For the second pressure sensor, variations in the slope and curvature of the curves are evident, which are not evident in the calibration curves of the first sensor in this example.

[0062] A white triangle shows the first measurement signal determined by the first pressure sensor on the p 1 -axis. At the same time, the second measurement signal, shown as a black triangle on the p 2 -axis, was determined by the second pressure sensor. Auxiliary lines emanating from the white triangle show which effective pressure would be expected depending on the type of gas in the shared measurement volume and which second measurement signal would be expected at this effective pressure. Gas G 3 best matches the actually measured values, so gas G 3 is defined as the result gas G*. The criterion for this can be, for example, distance on the - possibly logarithmic - p 2 -axis. The result pressure p* can now be read off the p eff -axis on the calibration curve for G* = G 3.

[0063] In the case where several such pairs of first and second measurement signals are to be compared, a sum of squared distances, for example, is a suitable criterion for determining the best-matching gas. The roles of the first and second measurement signals can be reversed in that, based on the actually measured second measurement signal, expected measurement signals are also determined on the p1 axis, and there—alternatively or additionally—the distance from the measured first measurement signal is determined as a criterion for the best-matching gas.

[0064] The method according to the invention and all its embodiments can be combined with an additional step of zeroing at least one of the pressure sensors. With various types of pressure sensors, the measurement signal generated for a specific effective pressure is subject to temporal drift. This effect can be eliminated by zeroing, which further increases the accuracy of the method according to the invention. The zero-point measurement signal is preferably determined at an effective pressure that is at least one to two decades below the measuring range of the pressure sensor to be zeroed. Checking whether a sufficiently low pressure is present can be done in various ways. For example, to zero a Pirani sensor, the achievement of a sufficiently low pressure can be checked using an ionization vacuum gauge whose pressure measuring range extends to at least two decades below the measuring range of the Pirani sensor.As another example, when zeroing a Bayard-Alpert ionization vacuum gauge, the achievement of a sufficiently low pressure can be verified by an extractor ionization vacuum gauge. Achieving a sufficiently low pressure for zeroing a pressure sensor can also be achieved, for example, through suitable process steps, such as prolonged pumping of the shared measurement volume. Alternatively, the achievement of a sufficiently low pressure can also be derived from the operating parameters of a vacuum pump operatively connected to the shared measurement volume of the pressure sensors.

[0065] It should be noted that pressure sensor drift and gas type dependence are two separate phenomena. For example, even if a pressure sensor is repeatedly zeroed after each gas type change, a gas type dependence remains.

[0066] Returning to the way in which a list of gas types, which may include a list of gas mixtures, can be handled, the following illustrative examples are given. For example, in the case of a Pirani sensor, the aim is to summarize the thermal conduction contributions of each component of the gas mixture to the total thermal conduction. For example, Jousten gives the following formula 11 from the publication K. Jousten, On the gas species dependence of Pirani vacuum gauges, Vac. Sci. Technol. A 26, 3, May / Jun 2008, gives a formula that takes into account the effective accommodation coefficients and heat capacities of each gas species involved in a mixture. Alternatively, and also suitable for Pirani sensors, Setiawan gives formulas 17 and 18 from the publication Ikhsan Setiawan et al., Critical Temperature Differences of a Standing Wave Thermoacoustic Prime Mover with Various Helium-Based Binary Mixture Working Gases, 2015 J. Phys.: Conf. Ser. 622 012010, which gives the thermal conductivity of the gas mixture based on the gas fraction, thermal conductivities, and molar masses of each gas species involved. Both formulas mentioned are suitable for generating a table of gas species in the form of gas mixtures, or for using a proportion of one or more gas species as a continuous parameter.The latter is advantageous for the "best fit" procedures described above.

[0067] A similar, albeit somewhat more complex, approach is possible with ionization vacuum gauges. Here, the energy distribution of the electrons, the ionization potential of the gases, the fragmentation of the gases, and any possible recombination are important factors for predicting an ion current. If the sensitivity S, which is defined as the ratio of the ion current at the ion collector on the one hand to the electron emission current and pressure on the other, is known from experimental data or simulations, the combined ion current can be determined as the sum weighted by the partial pressures of the participating gas species. Pressures and partial pressures are considered as the difference to the residual pressure, and collector currents are considered as the difference to the collector current at the residual pressure. List of reference symbols

[0068] 1 Group of pressure sensors 1' first pressure sensor 1" second pressure sensor 2 common measuring volume 3' first measuring signal output 3'' second measuring signal output 4' first pressure measuring range 4'' second pressure measuring range 5 means for storing calibration data 6 overlap pressure measuring range 12 control unit G* result gas type p 1 first measuring signal P 2 second measuring signal p eff effective pressure p* result pressure 100 Procedure 101 Step a) Providing calibration data 102 Step b) Acquiring the first and second measuring signals 103 Step c) Determining the result gas type 104 Step d) Determining the result pressure START Start of the procedure END End of the procedure

Claims

1. Method for operating a group (1) of pressure sensors (1', 1'') which are arranged to measure the pressure in a common measurement volume (2), wherein the group of pressure sensors comprises at least a first pressure sensor (1') having a first pressure measurement range (4') and a second pressure sensor (1'') having a second pressure measurement range (4''), wherein the first and second pressure measurement ranges overlap in an overlap pressure measurement range (6), wherein the first pressure sensor (1') is based on a first indirect pressure measurement principle and is adapted to output a first measurement signal calibrated to a reference gas (Gref), wherein the second pressure sensor (1'') is based on a second indirect pressure measurement principle and is adapted to output a second measurement signal calibrated to the reference gas, characterized in that the method comprises the steps of: a) providing (101) gas-type-specific first calibration data (K1[Gi]) for the first measurement signal and gas-type-specific second calibration data (K2[Gi]) for the second measurement signal, wherein the first and second calibration data describe a dependence of the first and second measurement signals, respectively, on the effective pressure (peff) and on a gas type in the common measurement volume for a list of gas types comprising at least one first gas type (G1) which is different from the reference gas; b) substantially simultaneously recording (102) a first measured value (p1) of the first measurement signal and recording a second measured value (p2) of the second measurement signal; c) determining (103) a resultant gas type (G*) as that gas type in the list of gas types which best matches the combination of the recorded first measured value (p1) and the recorded second measured value (p2), taking into account the first and second calibration data.

2. Method according to claim 1, comprising the additional step of d) determining (104) a resultant pressure (p*) as a function of the recorded first measured value (p1) and the first calibration data for the resultant gas type and / or as a function of the recorded second measured value (p2) and the second calibration data for the resultant gas type.

3. Method according to one of claims 1 or 2, wherein the first and second pressure sensors are vacuum pressure sensors.

4. Method according to one of claims 1 to 3, wherein the first pressure sensor (1') is a Pirani sensor.

5. Method according to one of claims 1 to 4, wherein the second pressure sensor (1'') is a hot-cathode ionization vacuum gauge, in particular of the Bayard-Alpert type.

6. Method according to one of claims 1 to 5, wherein the second pressure sensor (1'') is a cold cathode ionization vacuum gauge, in particular an inverted magnetron.

7. Method according to one of claims 1 to 6, wherein the first and second gas-type specific calibration data (K1[Gi], K2[Gi]) are each defined by a first and second factor (C1[Gi], C2[Gi]), respectively, by which the first measurement signal and the second measurement signal, respectively, are to be multiplied to obtain the effective pressure.

8. Method according to claim 7, wherein a list of quotients is formed by forming a quotient (Q[Gi]) for each gas type from the list from the first factor for the respective gas type and the second factor for the respective gas type, wherein a recorded quotient (Q) is formed as a quotient of the recorded first measured value (p1) and the recorded second measured value (p2), and wherein in step c) it is determined to which of the quotients (Q*) from the list of quotients the recorded quotient comes closest.

9. Method according to one of claims 1 to 8, wherein in step c) for each gas from the list of gases, starting from the recorded first measured value (p1), based on the first and second gas-type-specific calibration data, it is determined what value is expected for the second measurement signal if this gas were present in the common measurement volume, and wherein the smallest deviation of this value from the recorded second measured value (p2) is used as a criterion for determining the resultant gas type.

10. Method according to one of claims 1 to 9, wherein a plurality of pairs of a first measured value of the first pressure sensor and a second measured value of the second pressure sensor are each recorded as the pressure in the common measurement volume changes, wherein the gas type that best matches the combination of the recorded plurality of pairs is selected when determining the resultant gas type (G*).

11. Method according to one of claims 1 to 10, wherein it is checked on the basis of the recorded first measured value (p1), on the basis of the recorded second measured value (p2) or on the basis of the resultant pressure (p*) whether the pressure present in the common measurement volume lies in the overlap pressure measurement range and wherein the resultant pressure and / or the resultant gas type are rejected as invalid if this is not the case.

12. Apparatus (10) for carrying out the method according to one of claims 1 to 11, wherein the apparatus comprises a group (1) of pressure sensors arranged such that they are capable of measuring the pressure in a common measurement volume (2), wherein the group of pressure sensors comprises at least a first pressure sensor (1') having a first pressure measurement range (4') and a second pressure sensor (1'') having a second pressure measurement range (4''), wherein the first and second pressure measurement ranges overlap in an overlap pressure measurement range (6), wherein the first pressure sensor (1') is based on a first indirect pressure measurement principle and the second pressure sensor (1'') is based on a second indirect pressure measurement principle, and wherein the apparatus comprises means (5) for storing first calibration data and second calibration data, wherein the apparatus (10) comprises a control unit (12) which is operatively connected to a first measurement signal output (3') of the first pressure sensor, to a second measurement signal output (3'') of the second pressure sensor and to the means (5) for storing first calibration data and second calibration data for processing the measurement signals of the pressure sensors, characterized in that the control unit (12) is adapted to: i) output the resultant gas type, and / or ii) determine the resultant gas type and determine the resultant pressure based on the resultant gas type and output the resultant pressure.

13. Computer program product comprising instructions which, when the instructions are executed by a control unit (12) of an apparatus (10) according to claim 12, cause the control unit to perform the steps of a method (100) according to one of claims 1 to 11.