Determining the position of a first object relative to a second object

Abstract

A transmitter device (10) for determining the position of a first object relative to a second object, comprising at least one physical measure (14) connected to the first object with at least one code track (16), at least one scanning unit (18a-b) connected to the second object for generating at least one scanning signal by scanning the code track (16), an event memory, and a control and evaluation unit (22) configured to determine a first position signal in a first position channel by evaluating a scanning signal and a second position signal in a second position channel by evaluating a scanning signal, and to compare the first position signal and the second position signal with each other in order to detect a deviation, characterized in that the control and evaluation unit (22) is configured toOnly in the event of a deviation between the first and second position signals of at least one tolerance deviation, an event-based deviation event is stored in the event memory.

Description

[0001] The invention relates to a sensor device and a method for determining the position of a first object relative to a second object according to the preamble of claim 1 and 11 respectively.

[0002] Encoder devices are classified as either linear or rotary systems. A linear encoder determines a displacement along an axis. A rotary encoder, also known as an angle sensor or rotary position sensor, is used to detect an angle of rotation or the angular position of, for example, a shaft in a drive element. An important application is in motor feedback systems, where the rotary encoder in a servo motor reports the actual rotational speed back to the control system.

[0003] In the encoder device, a measuring scale is scanned, which is mounted along the linear axis or rotates with the shaft. The measuring scale acts as a modulator and generates an intensity modulation in a scanning sensor via its relative movement. Often, there is more than one scanned code track, or the existing code track is scanned multiple times. This can serve to acquire additional information, for example, to determine incremental and absolute positions. The commonly used sine / cosine encoders employ two offset scans to obtain a position.

[0004] Various technologies are used for the underlying sensor principles. The dimensional representation of optical encoders is a code disk or a corresponding elongated code body that rotates relative to a light source and the scanning sensor. For transmitted light in a transmission arrangement of light source and scanning sensor, optically transparent and opaque areas are provided; for reflected light in a reflection arrangement of light source and scanning sensor, reflective and absorbing areas are provided. These areas or code elements can be openings, reflective and diffractive structures, prisms, structures varying in transmittance, reflectivity, gray tone, or color, and other optically distinguishable elements. Together, the code elements form a code track or dimensional track.Magnetic encoders, for example, use a Hall sensor to detect the rotational movement of corresponding magnetic structures or encoding permanent magnets. Other physical principles are also suitable for position measurements; capacitive and inductive encoders are examples.

[0005] There is a growing demand for sensors that meet specific safety requirements. In safety engineering, stringent conditions are placed on the reliability and fail-safety of sensors in order to be approved, for example, in certain classes (SIL, Safety Integrity Level, or PL, Performance Level) of safety standards such as EN 61508. Typical safety functions include emergency shutdowns, shutdown of overheated equipment, and monitoring of hazardous movements. Devices intended to meet these requirements must be able to detect their own malfunction and, if necessary, initiate a safety-related response. However, the need for reliable measurement is not limited to safety engineering applications.

[0006] Numerous factors can limit the signal quality of a encoder device, causing deviations from the true position value and thus positional errors. Systematic influences that occur during the manufacturing or testing of the encoder device can often be adequately compensated for by adjustment. Alternatively, the encoder device must be rejected as scrap. In operation, an absolute measurement error is generally undetectable, as no reference system is available.

[0007] Relative measurement errors can conventionally be detected by cross-comparing the readings from two position channels of a sensor. Such a dual-channel design is a common measure to improve sensor reliability through redundancy or diverse redundancy. Diverse redundancy means that the two position channels are not simply duplicated, but differ in their design to increase robustness against a common source of error. If the deviation between the measured positions of the two position channels becomes too large, this is detected, and in applications subject to a safety standard, the system operated by the sensor is brought to a safe state. The relative measurement error itself cannot be corrected. If such an event occurs suddenly and without warning, the system operator faces an availability problem.In addition, the diagnosis is difficult because the reason for the deviation is unknown.

[0008] It is still common practice to try to detect influences leading to unacceptable measurement errors at an early stage (condition monitoring, predictive maintenance). However, in conventional encoder devices, this is only possible at best with complex methods that, for example, perform time-based or position-based frequency analysis. These methods are usually insufficient, especially for non-periodic, i.e., locally occurring, measurement errors caused by local contamination or damage to the measuring instrument. Even periodic measurement errors are difficult to detect and classify, as undersampling or aliasing can occur. Any more complex approaches are further hampered by the typically extremely limited computing resources within an encoder device.

[0009] Examples of the described multi-channel position determination include a multi-channel rotary encoder according to DE 10 2012 012 870 A1, a magnetically and optically diverse-redundant measuring device with two-channel scanning according to EP 2 113 742 B1 or a fail-safe monitoring of a speed encoder according to WO 2011 / 061345 A1.

[0010] EP 2 138 916 A1 discloses a position measuring system that, during operation, records state variables of the position measuring system or a servo drive and stores them in a memory. The position measuring system is not dual-channel. Furthermore, the complexity of a diagnosis is not reduced simply by storing state variables instead of evaluating them directly. This merely shifts the essentially same diagnosis to a later time or to a different device.

[0011] EP 2 527 868 A1 describes a safety laser scanner that records and classifies intrusions into protective fields. A sensor device uses a completely different measuring principle for other quantities, and intrusions into protective fields are not a category for a sensor device. Therefore, no diagnosis of a sensor device could be derived from a history of intrusions into protective fields.

[0012] DE 100 18 298 B4 discloses a method and a device for vibration detection in a position measuring device. A scanning device scans four graduation tracks of a scale and generates an absolute position measurement value from the scanning signals of the first three graduation tracks. Simultaneously, the increments swept over by the first and fourth graduation tracks are counted, and the difference in the counter readings is used to determine the current angular position of the scanning device relative to the scale. The extreme values ​​are stored as a measure of the temporal evolution of the angular positions, and a range of variation can be calculated from these stored extreme values. The extreme values ​​or the range of variation are displayed or transmitted to downstream electronics, and multiple values ​​of the range of variation for successive time periods can be displayed simultaneously on a screen.

[0013] EP 1 195 880 B1 presents a method for increasing the positional accuracy of an element movably arranged relative to a stator. Two Hall sensors each generate a sine and a cosine signal from a magnetic encoding. A distance between these two signals is determined, which, without positional errors, would correspond to the physical distance between the Hall sensors. Accordingly, deviations result in the positional error, which can thus be calibrated and subsequently compensated.

[0014] The object of the invention is therefore to improve fault detection in a sensor device.

[0015] This problem is solved by a encoder device and a method for determining the position of a first object relative to a second object according to claims 1 and 11, respectively. The two objects are movable relative to each other in a longitudinal or rotational direction, and the corresponding position or angular orientation is to be measured. In the case of linear motion, the encoder device is also referred to as a linear encoder; in the case of rotational motion, as a rotary encoder, angle sensor, or encoder; and when used in a servo motor, particularly as a motor feedback system. The first object is then preferably a shaft, and the second object is a stationary part or housing of the rotary encoder. Quantities such as velocity or angular velocity can be derived from the position or angular position.

[0016] The first object is a physical representation of scale with a scale track or code track, and the second object is a scanning unit that is thus movable relative to the physical representation. The terms "first object" and "second object" are interchangeable. The scanning unit generates at least one sample signal by scanning the code track. A control and evaluation unit evaluates this sample signal. The control and evaluation unit comprises, for example, one or more circuits or one or more digital processing units, such as at least one microprocessor, FPGA (Field Programmable Gate Array), or ASIC (Application-Specific Integrated Circuit).

[0017] The encoder device uses a two-channel position determination system with a first and second position channel. Position channels can be formed by multiple evaluations of the same sample signal and / or by evaluating multiple sample signals separately, and more than two position channels are possible. The first position signal from the first position channel and the second position signal from the second position channel are compared to detect any deviation. A minor deviation can still be tolerated. In safety-related applications, a significant deviation preferably triggers a safety-related response.

[0018] The invention is based on the fundamental idea of ​​treating a deviation between the two position signals as an event and recording such events. For this purpose, an event memory is provided, which the control and evaluation unit can access and in which a detected deviation event is stored.

[0019] The invention offers the advantage of significantly simplifying and enabling the early detection and classification of position measurement errors. Event-based storage of deviations results in considerably less data being generated. This allows for efficient evaluation even during operation and with limited computing resources in an encoder device. A time or position basis is unnecessary, thus avoiding aliasing or undersampling, which occur with conventional error monitoring. Overall, diagnostics are significantly simplified and improved, and the cause of errors can be much more easily identified from the deviation events. The availability of the system in which the encoder device is used is increased.

[0020] Preferably, two scanning units are provided, one assigned to the first position channel and the other to the second. In this embodiment, there are two physical position channels, each with its own scanning unit. The scanning units can scan the same code track or two different code tracks on the same or even different scales. It should be noted that a position determination can itself be based on multiple scans, as in the case of a sine / cosine encoder. Redundancy in two physical position channels only arises through an additional scan; that is, the scan signal for a position channel can certainly have multiple components such as sine / cosine. In this configuration, cross-channel embodiments are also conceivable, in which, for example, the cosine signal is used in both position channels and a separate sine signal is generated for each.As an alternative to two physical position channels with respective scanning units, dual-channel capability can also be achieved computationally by different evaluations of the scanning signal or the scanning signals of a common scanning unit.

[0021] The control and evaluation unit is designed to only record a deviation event if the first and second position signals differ by at least one tolerance deviation. Small differences between the position channels within this tolerance deviation are thus ignored. This makes the remaining deviation events more relevant, further reduces the amount of stored data, and simplifies its analysis.

[0022] The event memory preferably has an event table with a position value or position range and a corresponding event value. The event memory is structured by the event table, giving the stored deviation events a fixed format. The position value or position range indicates where the deviation event occurred, while the event value describes what happened. In particular, the event value represents the extent of the deviation, for example, by storing the difference between the two position values ​​as a numerical value or as an evaluative class such as "minor" to "critical." This structure is referred to as an event table, but other implementations are also included, each grouping a deviation event with its associated information in fixed categories, in this case, position value or position range and event value.The control and evaluation unit preferably uses the structure of the event memory; deviation events are stored and retrieved in the specified format as a position value or position range with the associated event value.

[0023] The control and evaluation unit is preferably designed to determine and store in the event memory the frequency with which a deviation event has occurred at a position value or within a position range. The position value corresponds to the position signal at the time of the deviation event. The first position signal, the second position signal, or a calculation such as their average can be used as the basis for this determination. A position range is an interval of positions around a position value, particularly symmetrical. According to this embodiment, if a deviation event occurs repeatedly at the same position value or within the same position range, a further deviation event is not stored each time. While this would be conceivable, it would waste storage space unnecessarily.Instead, additional information is stored for the already known deviation event, specifically a counter indicating the frequency of a deviation event at that position value or within that position range. This counter or frequency can be added to the event table described in the preceding paragraph as an additional column alongside the position value or position range and the event value.

[0024] The control and evaluation unit is preferably designed to store time information for each deviation event, indicating when the stored deviation event occurred. The deviation events thus receive a timestamp, enabling further diagnostics and evaluations. Like the counter or frequency, the time information can also form an additional column in an event table.

[0025] The control and evaluation unit is preferably designed to combine deviation events at adjacent position values ​​into a single deviation event within a position range, and / or to combine deviation events from a first position range into a larger second position range. This saves memory and further reduces the number of stored deviation events. This combining can occur with each new deviation event or only when memory becomes scarce, particularly when another deviation event could no longer be stored. Without combining, deviation events at directly adjacent or very close position values ​​would be stored separately. However, the added value of such deviation event information for diagnosis is often limited, as the underlying cause of the error is frequently the same.In particular, the frequency of deviation events in a summarized position environment is often just as valuable and easier to store and analyze. When summarizing, the summarized deviation events are preferably counted, and this frequency is stored as additional information for the summarized deviation event of the position range. If another deviation event occurs later in this position range, this deviation event is preferably no longer stored; it is ignored, or only the frequency is adjusted. If an initially selected first position range is relatively narrow, first position ranges can be summarized into larger second position ranges using the same principle by which position values ​​are summarized into a position range.In principle, this could be extended to even larger position ranges if needed, but at some point there is no longer enough position resolution to meaningfully evaluate the deviation events.

[0026] The control and evaluation unit is preferably designed to prioritize deviation events, particularly when the event memory exceeds a storage limit, and to store a new deviation event only if it is more important than an earlier deviation event it overwrites, and / or to delete at least one least important deviation event from the event memory. This allows the most important deviation events to be retained even with limited event memory. A new deviation event is preferably only stored and thus considered subsequently if, according to its prioritization, it is more important than the deviation events already stored, and it can then displace a previously stored, less important deviation event, which is deleted. It is possible to begin prioritization only after a certain portion of the memory has already been filled.In particular, the event memory can be cleaned up cyclically or when a storage limit is exceeded, according to a prioritization, whereby only the most important deviation events are retained and the rest are deleted. The storage limit can be the physical limit of the event memory or artificially set, for example, for a fixed maximum number of deviation events to be stored.

[0027] The control and evaluation unit is preferably designed to prioritize based on event values. The event value is, in particular, a difference between the position signals or an evaluation of the deviation event. Prioritization based on event values ​​therefore means giving precedence to more significant deviation events. For example, if there is already a deviation event at a position value or within that position range, it will only be overwritten by a more significant deviation event and otherwise remain unchanged. Alternatively, the frequency of deviation events at that position value or within that position range will be tallied, and only the event with the strongest contributing value will be retained.

[0028] The control and evaluation unit is preferably designed to prioritize deviation events according to their age. For example, more recent deviation events can take precedence, or the oldest deviation event can be retained to track how long a deviation has existed.

[0029] The control and evaluation unit is primarily designed to prioritize deviation events based on their frequency. This prioritization is particularly relevant when a deviation event currently being evaluated occurs at a new position value or within a new position range, as otherwise the frequency of the earlier deviation event could be inflated. In this case, the new deviation event, which is occurring for the first time at this position value or within this position range, cannot prevail based on its frequency alone. However, it may be advisable to eliminate another individual event instead, in order to avoid losing the most recent deviation event.

[0030] The control and evaluation unit is preferably designed to prioritize based on a combination of event values, age, and / or frequency. These three prioritization criteria can be applied individually or in combination. For example, a recent, significant, and recurring anomaly takes precedence over an outdated, minor, and isolated anomaly. Other combinations are also possible.

[0031] The control and evaluation unit is preferably designed to remember the most recently stored deviation event and only consider a subsequent deviation event if its position value differs from that of the last stored deviation event by a minimum amount. In other words, it is required that there has been a certain minimum movement, and thus a minimum change in position, before a further deviation event is stored. This prevents the physically identical deviation event from being stored multiple times in the case of very slow movement or a standstill, and thus potentially filling up the event memory.

[0032] The control and evaluation unit is preferably designed to assess the deviation events collected in the event memory and to issue a failure warning or a safety-related signal if the number and / or extent of the deviation events exceed a functional tolerance threshold. This integrates diagnostics directly into the sensor device, and this is made possible even with limited evaluation resources by the compressed and highly informative storage of deviation events according to the invention. This allows for automatic and timely warnings before a failure. It is also conceivable that the evaluation might indicate that further safe operation is not possible and therefore trigger a safety-related response in the system, even though each deviation event, individually, might not have been safety-critical.One evaluation criterion might be that there are too many and / or too severe disturbances, which can be attributed to the number or frequency of disturbances and their event severity. These two factors can be interdependent; for example, a large number of minor disturbances might be tolerated, as might a single major disturbance, but not a large number of disturbances, including some that are more severe. The evaluation can help identify a cause, such as a local error due to the absence, damage, or contamination of part of the code track or measure, or a periodic error resulting from imbalance or eccentricity of the measure.

[0033] The method according to the invention runs automatically and is therefore, in particular, computer-implemented, preferably in a encoder device. The method can be further developed in a similar manner to the encoder device itself and exhibits similar advantages. Such advantageous features are described by way of example, but not exhaustively, in the dependent claims following the independent claims.

[0034] The invention is further explained below with regard to additional features and advantages by way of example embodiments and with reference to the accompanying drawing. The illustrations in the drawing show: Fig. 1 a schematic representation of a sensor device; Fig. 2 an exemplary flowchart of the cross-comparison of two position channels with storage of deviation events; Fig. 3. An example event table for deviation events; Fig. 4. A graphical representation of deviation events at individual position values; and Fig. 5 a graphical representation of deviation events in position areas.

[0035] Fig. Figure 1 shows a schematic representation of a encoder device 10 in an embodiment as an encoder or rotary encoder. It is used, for example, in a motor feedback system. The encoder device 10 has a preferably circular code disk or scale 14 rotating with a shaft 12, and a scale track or code track 16 is located on the scale 14. Two scanning units 18a-b are arranged on the circumference of the scale 14 or the code track 16. Each of these scanning units 18a-b scans the code track 16 during the rotational movement with one or more individual scanning units (not shown here) or, specifically in this optical embodiment, with photodiodes. The individual scanning units are preferably integrated on an opto-ASIC (application-specific integrated circuit) and receive the light from a light source 20a-b associated with the scanning units 18a-b, which penetrates the code track 16.

[0036] A control and evaluation unit 22 evaluates the sampling signals from the sampling units 18a-b to determine the angular position of the scale 14. The control and evaluation unit 22 comprises one or more components, such as a microcontroller, an ASIC (Application-Specific Gate Array), or an FPGA (Field Programmable Gate Array). During evaluation, the sampling signals from the two sampling units 18a-b are used for dual-channel position determination. In other embodiments, the sampling units 18a-b can each sample their own code tracks or even their own scales for a dual-channel setup. Conversely, dual-channel functionality can also be achieved with only a single sampling unit 18a-b and different evaluation methods.The type of scanning signals, thus the design of code track 16 and scanning units 18a-b, as well as their specific evaluation, can be varied; the invention does not specify in any particular way how scanning signals are detected and thus an angular position is measured.

[0037] The angular position or quantities derived from it, such as velocity information after differentiating the angular position, can be provided at an output. Preferably, the evaluation takes place digitally after appropriate analog-to-digital conversion. In principle, the control and evaluation unit 22 can also be provided, at least partially, within the sampling units 18a-b or externally; in the latter case, the encoder device 10 outputs, for example, the sampling signals and / or intermediate results such as a reconstructed modulation of the duty cycle or incremental angular positions for further processing.

[0038] The representation of the encoder device 10 in Fig. Figure 1 is very schematic. Therefore, the design of the scanning units 18a-b with associated light emitters 20a-b is very simple. In particular, with regard to the dimensions and specific positions of the components, the encoder device 10 can vary considerably in practice. Fig. 1. As an alternative to a transmissive design, a reflective design is also conceivable, in which scanning units 18a-b and light emitters 20a-b are located on the same side of the scale 14 and the code track 16 has reflective properties. An embodiment as an optoelectronic encoder device 10 is also exemplary; alternatively, according to the invention, a magnetic, inductive, or capacitive encoder device 10 can be provided, or another suitable physical operating principle for detecting scanning signals can be used. In a magnetic system, for example, instead of transparent and opaque areas of the code track 16, north poles and south poles alternate, which are detected by Hall sensors or the like of the scanning units 18a-b. Reference is made here to the introductory explanations regarding sensor, measurement, and position determination principles of encoder devices.

[0039] The encoder device 10 can also be configured as a linear encoder or longitudinal measuring system instead of a rotary encoder. In this case, the scale 14 is not circular but elongated, the code track is arranged accordingly on a straight line instead of a circle, and the scanning units 18a-b are arranged on this elongated code track.

[0040] Fig. Figure 2 shows an example flowchart for dual-channel position determination with subsequent cross-comparison. Thanks to its dual-channel design, the encoder device 10 can be configured as a safe encoder device 10 in accordance with safety standards such as EN 61508.

[0041] In step S1, a first position signal is determined from the sampling signals of the scanning units 18a-b in a first position channel. Similarly, in step S2, a second position signal is determined from the sampling signals of the scanning units 18a-b in a second position channel. These two steps can be performed sequentially or in parallel. It is possible to assign one of the two scanning units 18a-b to one position channel and the other to the other position channel. As mentioned several times already, multiple position determinations can also be performed in other ways based on the sampling signals and even with only one scanning unit. The respective position value output by the encoder device 10 can be the current first position signal, the current second position signal, or a calculation such as an average of the two position signals.

[0042] In step S3, the two position signals are compared. In step S4, it is checked whether a significant deviation occurs. A deviation is significant, for example, if the difference between the two position signals exceeds a tolerance deviation. If this is the case, it is stored as a deviation event in an event log in step S5. Thus, deviations between the two position channels are detected on an event-based basis, and the previously occurring deviation events are available in the event log for diagnostic purposes. After step S4 or S5, the next position determination cycle begins.

[0043] Fig. Figure 3 shows an example of the structure of an event table for deviation events. Deviation events can be stored in the event memory using this structure. Each row represents a deviation event. The first column records the position value at which the deviation event occurred. This preferably corresponds to the position value from one of the two position channels or a combination of the position channels, which also serves as the output value. In other embodiments, a position range is recorded here, i.e., a neighborhood of position values. An optional second column records an event value that characterizes the deviation event. This can be, in particular, the difference between the two position signals or an evaluative category such as "minor" to "critical".

[0044] An optional third column can store a frequency value for each deviation event. In this case, deviation events are preferably not simply written sequentially to new rows, but rather a check is performed to see if a previous deviation event is already stored at the current deviation event's position. If so, its frequency is incremented; otherwise, the new deviation event is appended to the event table in a new row. A further optional fourth column can store a timestamp for each deviation event.

[0045] In a simple implementation, the deviation events are simply written sequentially into new lines. More intelligent rules are conceivable in various implementations to further increase the informative value of the stored deviation events or to manage with the fewest possible stored deviation events. This reduces the data and simplifies further analysis; it also allows the event memory to remain small. A first such rule has already been described, namely that deviation events are counted at the same positions and not stored individually.

[0046] These and the following additional rules are optional and can also be combined. In a standstill monitoring system that can also take slow movement into account, the control and evaluation unit 22 always remembers the last stored deviation event. A new deviation event is only considered if its position differs from the last stored deviation event by a minimum distance. This prevents the event memory from rapidly filling up with essentially the same deviation event when the system is stationary or moving slowly.

[0047] This is also achieved if, instead of a single position value, a certain position range, preferably symmetrical around the position value, is stored for a deviation event. A rule then dictates that only one deviation event is stored for each position range. For a new deviation event, it is first checked whether it falls within the position range of a previous deviation event. If so, the previous deviation event is overwritten or its frequency is increased, and the event value may be modified, for example, to be the maximum of all previous event values ​​in that position range. The position range can also shift in the direction of the newly added deviation event. A deviation event in a position range that does not yet exist in the event memory is stored separately.

[0048] Furthermore, it is possible to combine multiple deviation events with position values ​​into a single deviation event within a common position range. Event values ​​can be combined, for example, to determine a maximum, and a cumulative frequency can be stored. Similarly, multiple deviation events with position ranges can be combined into a single deviation event with a larger position range.

[0049] If the event memory is already full, which may be due to a physical storage limit or a desired maximum number of stored deviation events, there are additional optional rules for deleting some deviation events from the event memory. Prioritization is preferably based on event value, frequency, and / or time. The possibility of overwriting a deviation event at a position for which a deviation event was already stored has already been mentioned. Furthermore, a lower-priority, earlier deviation event can be deleted to make way for a new deviation event. Also mentioned is the ability to combine position values ​​into position ranges or position ranges into larger position ranges. This also frees up a portion of the event memory. The possible sizes of position ranges can be taken from a predefined selection or chosen dynamically.

[0050] Fig. Figure 4 illustrates some stored deviation events with their position values ​​on the X-axis and the associated event values ​​on the Y-axis. Fig. Figure 5 is a corresponding illustration based on position ranges. The respective local distributions can have resulted from multiple deviation events. However, it is also possible to assign a decreasing event value to the neighboring positions of a deviation event. A deviation event or multiple deviation events are thus distributed across a certain neighborhood. The representations of the Fig. 4 and Fig.The underlying position and event information is automatically generated through the described method of event-based storage of deviations between position channels. This results in dynamic accumulations in local position areas. Only a few stored deviation events, or a few rows in an event table, are required for this.

[0051] The deviation events collected in this way can be evaluated continuously or through targeted queries. The position, extent, and frequency of the deviation events are directly accessible in the event memory. Since the geometric arrangement of the position channels or the scanning units 18a-b that feed the position channels is also known, it is very easy to identify local defects, damage, or contamination of the code track 16 or the scale 14. Likewise, any periodically occurring deviation events can be detected. These could, for example, relate to the rotation of the scale 14, the periodicity of rotating balls in a ball bearing, or the influence of a gear tooth. As a result of the evaluation, warnings can be issued in a timely manner, indicating a suspected cause, in order to prevent a sudden failure.

[0052] The sign of the deviation events allows conclusions to be drawn about the position channel that caused the deviation event. Knowing a geometric distance between the position channels, in particular an offset between the scanning units 18a-b, this can be used to check the plausibility of deviation events, since the positional distance between two deviation events with the same cause corresponds to the positional distance between the position channels. The two position channels can generate their position signals inverted relative to each other.

[0053] Thanks to the compressed storage in the form of a few deviation events, which, when the rules described above are applied intelligently, also contain particularly relevant information, highly efficient logging and analysis are possible. The stored position values ​​each have a significantly higher resolution than would be possible with the number of stored deviation events or table rows in conventional time-based or position-based storage.

Claims

[1] A transmitter device (10) for determining the position of a first object relative to a second object, comprising at least one physical measure (14) connected to the first object with at least one code track (16), at least one scanning unit (18a-b) connected to the second object for generating at least one scanning signal by scanning the code track (16), an event memory, and a control and evaluation unit (22) configured to determine a first position signal in a first position channel by evaluating a scanning signal and a second position signal in a second position channel by evaluating a scanning signal, and to compare the first position signal and the second position signal with each other in order to detect a deviation, characterized by, that the control and evaluation unit (22) is designed to store a deviation event in the event memory only in the event of a deviation between the first and second position signal by at least one tolerance deviation. [2] Encoder device (10) according to claim 1, wherein two scanning units (18a-b) are provided and wherein one scanning unit (18a-b) is assigned to the first position channel and the other scanning unit (18b-a) is assigned to the second position channel. [3] Transmitter device (10) according to one of the preceding claims, wherein the event memory has an event table with each position value or position range and an associated event value. [4] Encoder device (10) according to one of the preceding claims, wherein the control and evaluation unit (22) is designed to determine and store in the event memory the frequency of how often a deviation event has occurred at a position value or in a position range. [5] Transmitter device (10) according to one of the preceding claims, wherein the control and evaluation unit (22) is configured to store time information for a deviation event, indicating when the stored deviation event occurred. [6] Encoder device (10) according to one of the preceding claims, wherein the control and evaluation unit (22) is configured to combine deviation events at adjacent position values ​​into a deviation event in a position range and / or deviation events in a first position range into a larger second position range. [7] Transmitter device (10) according to one of the preceding claims, wherein the control and evaluation unit (22) is designed to prioritize deviation events, in particular when the event memory exceeds a storage limit, and to store a new deviation event only if it is more important than an earlier deviation event overwritten by it and / or to delete at least one least important deviation event in the event memory. [8] Transmitter device (10) according to claim 7, wherein the control and evaluation unit (22) is configured to prioritize according to event values, age of the deviation events and / or frequency of the deviation events. [9] Transmitter device (10) according to one of the preceding claims, wherein the control and evaluation unit (22) is designed to remember each last stored deviation event and to take into account a further deviation event only when its position value differs by a minimum distance from that of the last stored deviation event. [10] Transmitter device (10) according to one of the preceding claims, wherein the control and evaluation unit (22) is designed to evaluate the deviation events collected in the event memory and to issue a failure warning or a safety-related signal when the number and / or extent of the deviation events exceed a functional tolerance threshold. [11] Method for determining the position of a first object relative to a second object, wherein a measure (14) with at least one code track (16) is connected to the first object, which is sampled by at least one scanning unit (18a-b) connected to the second object in order to generate at least one sampling signal, wherein a first position signal in a first position channel is determined by evaluating a sampling signal and a second position signal in a second position channel is determined by evaluating a sampling signal, and the first position signal and the second position signal are compared with each other to detect a deviation, characterized by , that only in the event of a deviation between the first and second position signal by at least one tolerance deviation, an event-based deviation event is stored in an event memory.

Images