Position measuring arrangement

Applicants claim, under 35 U.S.C. § 119, the benefit of priority of the filing date of Feb. 15, 2006 of a German patent application, copy attached, Serial Number 10 2006 007 184.0, filed on the aforementioned date, the entire contents of which is incorporated herein by reference.

1. Field of the Invention

The present invention relates to a position measuring arrangement for the determination of an absolute position. The present invention furthermore relates to a method for operating an absolute position measuring arrangement.

2. Discussion of Related Art

Position measuring arrangements in the form of angle- and length-measuring arrangements are employed in the machine tool industry and other manufacturing, handling and testing systems. A high functional dependability becomes of increased importance in connection with all uses, because faulty operation can cause considerable damage.

Absolute position measuring arrangements, which can provide correct position information in any relative position, even immediately following an interruption in the energy supply, are being increasingly employed for position measuring. In this case, the absolute position is represented by a single-track code arranged in a particularly space-saving manner of code elements arranged one behind the other in the measuring direction. Here, the code elements are arranged behind each other in a pseudo-random manner, so that a definite number of successive code elements respectively constitute a code pattern or bit pattern, which definitely defines the absolute position as a code word. A new code word is already formed in the course of the displacement of the scanning device by a single code element, and a sequence of different code words is available over the entire measuring range to be recorded. Such a sequential code is called a chain code or pseudo-random code.

A number of measures for assuring the dependable read-out of the code elements have become known in the prior art. These measures have in common that it is intended to assure that the code elements are scanned within the definite range, i.e. not in the transition area to the adjacent code elements. A selection information is obtained for this from an auxiliary track, or from the code track itself, by which the scanning elements are selected for assured scanning and generation of a code word. The binary information from the selected scanning elements is provided to a decoding device for forming the instantaneous absolute position. Stored assignment tables or generators are used as decoding device.

Such position measuring arrangements and methods are described, for example, in DE 42 09 629 A1, DE 39 42 625 A1, DE 43 09 863 C1, DE 38 25 097 C2 and WO 03/060431 A1.

It is customary to employ redundant scanning units for increasing the operational dependability and for detecting faulty operation of absolute position measuring arrangements. Such a position measuring arrangement is described in EP 0 789 226 B1. Several spaced apart code words are scanned and compared with each other. The hardware outlay is increased by this multiple redundancy, and connected therewith as the costs as well.

It is therefore an object of the present invention to dependably discover a faulty operation of the measuring arrangement by as low as possible an outlay for hardware.

This object is attained by a position measuring arrangement that includes an arrangement of scanning elements for scanning an absolute code and a selection system for selecting scanning signals from the arrangement of scanning elements by a first method of selection and a second method of selection different from the first method of selection. The position measuring arrangement further includes a decoding device for forming a first absolute measured position value from the scanning signals selected by the selection system when the first method of selection is applied to the selection system and a second absolute measured position value from the scanning signals selected by the selection system when the second method of selection is applied.

This object is also attained by the method for operating a position measuring arrangement that includes scanning an absolute code by an arrangement of scanning elements and selecting scanning signals from the arrangement of scanning elements by a first method of selection and a second method of selection different from the first method of selection. The method includes forming a first absolute measured position value from the scanning signals selected in accordance with the first method and forming a second absolute measured position value from the scanning signals selected in accordance with the second method. The method includes making the first absolute measured position value and the second absolute measured position value available for error checking.

An advantage of the present invention lies in particular in that it is possible in a simple manner to test the freedom from errors of the scanned code words, or generated measured position values and that, in case of the appearance of an error, it is possible to change into a dependable state of a drive unit whose absolute position is determined by the position measuring arrangement.

By the diversity of the selection function of scanning elements it is possible in an advantageous manner to use the hardware, which is provided for dependable scanning anyway, which minimizes costs and structural size.

Further characteristics and advantages of the present invention will become apparent in the course of the following description of an exemplary embodiment by the drawings.

FIG. 1 is a first embodiment of an absolute position measuring arrangement in accordance with the present invention;

FIG. 2 shows a first embodiment of a position measuring arrangement to be used with the absolute position measuring arrangement of FIG. 1 in accordance with the present invention;

FIG. 3 shows a second embodiment of a position measuring arrangement to be used with the absolute position measuring arrangement of FIG. 1 in accordance with the present invention;

FIG. 4 represents the scanning of a code in detail using the position measuring arrangements of FIGS. 1 and 2 ; and

FIG. 5 shows a signal diagram associated with the position measuring arrangements of FIGS. 1 , 2 and 4 .

An absolute position measuring arrangement, designed in accordance with the present invention, is schematically represented in FIG. 1 . This position measuring arrangement is a linear measuring arrangement, however, the present invention can also be employed in connection with angle-measuring arrangements.

For position measuring, a code C of a scale 1 is arranged to be movable with respect to a scanning device 2 . For measuring lengths, the scale 1 is embodied in a rod or tape shape, and for measuring angles the scale is embodied in the shape of a drum or disk.

The code C is arranged on a track and includes code elements C 1 to C 5 , which follow each other in the measuring direction X. These code elements C 1 to C 5 form a chain code, i.e. they are distributed pseudo-randomly in the measuring direction X and form a continuous sequence of different code words. In FIGS. 1 and 2 the code elements C 1 to C 5 are identified by 0 or 1. This identification symbolically characterizes the different physical properties of the individual code elements C 1 to C 5 . If this is a code C which can be photo-electrically scanned, the code elements C 1 , C 4 identified by 0 are not transparent, and the code elements C 2 , C 3 , C 5 are transparent, or the code elements C 1 , C 4 identified by 0 are non-reflecting and the code elements C 2 , C 3 , C 5 are reflecting. The differing characteristics 0 and 1 can also be generated by the sequence of partial areas within a code element C 1 to C 5 , as disclosed in WO 03/060431 A1, which will be explained later in greater detail by an example.

The scanning device 2 contains an arrangement of scanning elements D 1 to D 6 for scanning the code C. Several scanning elements D 1 to D 6 are respectively assigned to each code element C 1 to C 5 , which means that the mean distance between the scanning elements D 1 to D 6 is a fraction of the length of a code element C 1 to C 5 . The scanning signals S 1 to S 6 of the scanning elements D 1 to D 6 are supplied to a selecting system that can include a first selecting device 3 and a second selecting device 4 . On the basis of first selection information A 1 , the selecting device 3 selects some of the scanning signals S 1 to S 6 for further processing and for forming a code word CW 1 . This code word CW 1 is supplied to a decoding device 5 for forming a first absolute position POS 1 . Note that the selection information A 1 can be thought of as a software instruction or switch control signal that controls which of the signals S 1 to S 6 are supplied to selecting device 3 .

The scanning signals S 1 to S 6 of the scanning elements D 1 to D 6 are also supplied to the second selecting device 4 . On the basis of a second selection information A 2 , this second selecting device 4 selects some of the scanning signals S 1 to S 6 for further processing and forming a code word CW 2 . This code word CW 2 is supplied to the decoding device 5 , or to a separate coding device (not represented), for forming a second absolute position POS 2 . Like selection information A 1 , selection information A 2 can be thought of as a software instruction or switch control signal that controls which of the signals S 1 to S 6 are supplied to selecting device 4 .

Decoding takes place either by a stored assignment table in which the corresponding measured position value POS 1 , POS 2 is assigned to code words CW 1 and CW 2 , respectively, or by a generator, which generates the code word sequence by an accompanying counter and which, when the generated code word agrees with the scanned code word CW, issues the setting of the counter as a measurement of the position. The latter version can be realized by a shift register.

The selection information A 1 and A 2 is determined in accordance with different criteria, therefore the selection takes place diversely. By the diversity of the selection it is assured that faulty scanning is detected even in the absence of complete hardware redundancy of the scanning process. For detecting this error, in the simplest case the two positions POS 1 and POS 2 are checked for equality, and an error signal is generated in case of inequality.

The selection of scanning signals S 1 to S 6 assures that only definite scanning signals, such as S 2 , S 4 , S 6 , are used for forming the code words CW 1 , CW 2 in each relative position of the scale 1 and the scanning device 2 . In the example in accordance with FIG. 1 , definite scanning signals S 2 , S 4 , S 6 are only generated by the scanning elements D 2 , D 4 , D 6 , which clearly respectively scan only a code element C 2 , C 3 , C 4 , i.e. the mean range of a code element C 1 to C 5 . It is assured by the selection that uncertain scanning signals S 1 , S 3 , S 5 are not used for further processing, i.e. for position determination. Those scanning signals S 1 , S 3 , S 5 are uncertain, which are generated by the scanning elements D 1 , D 3 , D 5 which scan the transition areas between respectively two code elements D 1 to D 5 which follow each other, i.e. are simultaneously affected by the physical state of two code elements C 1 to C 5 .

A particularly advantageous further processing of the first absolute position POS 1 and the second absolute position POS 2 will be explained by FIG. 1 . In order to also include the interface 9 , in particular a serial one, and the transmission between the position measuring arrangement and a follow-up unit 10 in the test for correctness, both measured absolute position values POS 1 and POS 2 are transmitted to the follow-up unit 10 , for example a numeric control, or a drive unit, and the comparison for the purpose of checking for errors takes place only then.

In this case it is advantageous if at least one of the measured absolute position values POS 1 and POS 2 is altered prior to transmission in such a way that the two values transmitted via the interface 9 are different. In the simplest case a known offset OF is imposed on one of the measured absolute position values POS 1 , POS 2 in a changing unit 8 , and can then again be taken into consideration during the comparison in the follow-up unit 10 . The offset OF is an arbitrarily determined by a known algorithm, wherein the follow-up unit 10 is aware of the algorithm used. The offset OF is imposed on one of the absolute position values to ensure that the position values transmitted are different and to heighten the chance that any errors caused by transmission of the two values are detected by the follow-up unit 10 .

In many cases, the resolution of the absolute position measurement by the code C alone is not sufficient for a highly accurate position measurement. For this reason several possibilities exist for complementing the absolute position POS 1 with a higher-resolving measured position value POS 3 . This higher-resolving measured position value POS 3 can be obtained in different ways, for example directly from the code C itself, in which the position of the edges of the code elements C 1 to C 5 (transitions) in regard to the scanning elements D 1 to D 6 is evaluated in addition, and a periodic incremental signal is obtained from this, which is interpolated in a known manner. A further possibility for obtaining a higher-resolving measured position value POS 3 includes using additional information Z is arranged parallel with the code C, for example in the form of one or several incremental tracks ( FIG. 2 ).

This higher-resolving measured position value POS 3 is linked with at least one of the absolute measured position values POS 1 , POS 2 , and the resultant absolute measured position value POS 4 obtained from this, charged with the offset OF as a value (POS 1 +POS 3 +OF), is transmitted to the follow-up unit 10 . The resultant absolute measured position value POS 4 is composed of a sequence of bits, of which the lesser value bits obtained by the higher-resolving position measurement further resolve the absolute measured position value POS 1 . If the higher-resolving measured position value POS 3 is only linked with one of the absolute measured position values POS 1 , only the bits with the same resolution are compared with each other. For example, suppose POS 1 =010, POS 2 =010, POS 3 =011 and POS 4 =POS 1 +POS 3 =010011. When comparing POS 4 with POS 2 only the first three bits of POS 4 (010) are compared with the three bits 010 of POS 2 .

Possibilities for forming the selection information A 1 and A 2 will now be explained in what follows.

A first exemplary embodiment for generating the selection information A 1 , A 2 will be explained in greater detail by FIG. 2 . In this example the scale 1 has also a piece of additional information Z, besides the code C. This additional information Z on the scale 1 is arranged parallel with the code C, and by scanning the additional information Z with detector elements E 1 , E 2 a position POS 3 is obtained from it, which divides a code element C 1 to C 5 into several sections, i.e. definitely defines the absolute position POS 3 within a code element C 1 to C 5 . This additional information is, for example, an incremental graduation with a graduation period corresponding to the width of one code element C 1 to C 5 . Scanning signals K 1 , K 2 , which are phase-shifted with respect to each other, are generated by scanning the incremental graduation having the additional information Z via detector elements E 1 , E 2 , from which the absolute position POS 3 within a graduation period is calculated in an interpolation unit 6 . Because the position of the incremental graduation is fixedly assigned to the code elements C 1 to C 5 , it is therefore possible to determine the exact position of the scanning device 2 with respect to the scale 1 , and therefore the exact position of the scanning elements D 1 to D 6 with respect to the code elements C 1 to C 5 , and this absolutely definitely within a respective code element C 1 to C 5 . Based on receiving absolute position POS 3 , selection device 3 selects only those signals S 1 to S 6 which are derived from the center of code elements C 1 to C 4 to be processed. Thus, POS 3 is the selection information A 1 . On the one hand, this absolute position POS 3 complements the position value POS 1 in that a resultant absolute measured position value POS 4 is formed in a logical connecting device 7 , known per se, and on the other hand it constitutes the selection information A 1 . Such a selection method is also described in DE 39 42 625 A1 and DE 42 09 629 A1.

In accordance with the present invention, a further selection is now performed diversely. In the example in accordance with FIG. 2 , the further selection is performed with the aid of the code C itself. This is graphically explained by FIGS. 4 and 5 . The code elements C 1 , C 2 , C 3 (only three are shown for reasons of space) each includes two partial areas A and B of equal length, which are arranged following each other immediately in the measuring direction X. The partial areas A and B of a code element C 1 , C 2 , C 3 are embodied complementary to each other, this means that they have inverse properties, i.e. in accordance with the optical scanning principle of FIG. 4 they are transparent and non-transparent, or in case of incident light scanning, are reflecting or non-reflecting.

In order to generate a digital value, or a bit, for each code element C 1 to C 3 , the scanning signals S 1 to S 12 of the two partial areas A, B of a respective code element C 1 to C 3 are compared with each other. In case of a displacement of the scanning device 2 with respect to the code C by the width, or the length, of a code element C 1 , C 2 , C 3 , a new code word is generated, and a multitude of different code words is formed over the measuring range which is to be absolutely measured.

FIG. 4 shows an instantaneous position of the code C in relation to the scanning device 2 . The scanning elements D 1 to D 12 are arranged in sequence at a mean distance of half the width of a partial area A, B of the code C. It is assured by this that in every position at least one scanning element D 1 to D 12 is definitely assigned to a partial area A, B of a code element C 1 , C 2 , C 3 and does not scan a transition between two partial areas A, B. In the position represented, the partial area C 1 , A is scanned by the scanning element D 1 and the partial area C 1 , B by the scanning element D 3 . The scanning elements D 1 , D 3 detect the light distribution and, as a function of the light intensity, generate an analog scanning signal S 1 , S 3 proportionally to the light intensity. Since the two partial areas C 1 , A and C 1 , B are embodied complementary to each other, the light intensity of the scanning signals S 1 and S 3 is inverse with respect to each other, therefore the signal levels are spaced far apart from each other.

This signal distance is now used for generating the binary information B 1 by checking which of the two scanning signals S 1 , S 3 of the code element C 1 is stronger. This check can be performed by forming a quotient or by forming a difference. Difference formation is employed in connection with the example, wherein in accordance with FIG. 4 a trigger module T 1 is used as the comparison device. The trigger module T 1 generates B 1 =0, when S 1 is less than S 3 , and B 1 =1, when S 1 is greater than S 3 . Binary information B 2 to B 6 is obtained in the same way by scanning the code elements C 2 , C 3 and comparing the analog scanning signals S 1 to S 12 by means of trigger modules T 1 to T 6 . For reasons of space, not all comparison elements (trigger modules) have been represented in FIG. 4 , therefore in actuality the scanning signals S 3 of the scanning element D 3 are compared with the scanning signal S 5 of the scanning element D 7 , S 4 with S 6 , S 7 with S 9 , S 8 with S 10 , etc.

Therefore a first digital value is assigned to a first sequence of the partial areas A, B, designed to complement each other, and a second digital value to a second sequence of the partial areas A, B, designed to complement each other. In the example, the value 0 is assigned to the sequence of opaque—transparent, and the value 1 to the sequence transparent-opaque.

Since the two partial areas A, B of each code element C 1 , C 2 , C 3 complement each other, the interference distance between the scanning signals S 1 to S 12 is very large. A change in the light intensity of the light source L affects the scanning signals S 1 to S 12 of both partial areas A, B equally.

Based on the complementary design of each of the two partial areas A, B of a code element C 1 , C 2 , C 3 , respective analog scanning signals S 1 to S 12 , whose difference exceeds a predetermined value, must be generated by scanning these partial areas A, B in the course of the correct operation of the position measuring arrangement. A satisfactory error check is possible by observing this difference value. The basis of this error check is that it can be assumed that, when the difference value falls below a predetermined amount, the binary information B 1 to B 6 is uncertain, so that therefore an error signal F is generated with respect to this binary information B 1 to B 6 .

By forming the difference (S 1 −S 3 ) of the analog scanning signals S 1 and S 3 of the code element C 1 a check is made whether the difference amount exceeds, or does not exceed, a predetermined comparison value V. If the difference amount (S 1 −S 3 ) does not exceed the predetermined comparison value V, an error signal F is issued. These signal conditions are represented in FIG. 5 .

The arrangement of the two partial areas A, B of each code element C 1 , C 2 , C 3 which follow each other directly side-by-side in the measuring direction X has the advantage that the scanning elements D 1 to D 12 can be arranged side-by-side at a short distance in the measuring direction X, so that therefore the position measuring arrangement is insensitive to twisting of the scanning device 2 with respect to the code C, i.e. to Moiré fluctuations.

Furthermore, sensitivity to dirt is low, since it can be assumed that both partial areas A and B of a code element C 1 , C 2 , C 3 would be equally affected.

It can easily be seen in connection with the example of the scanning elements D 1 and D 2 in FIG. 4 that in the instantaneous position represented the even-numbered scanning elements D 2 , D 4 , D 6 , D 8 , D 10 , D 12 are respectively located at a transition between two partial areas A, B, and therefore the trigger modules T 2 , T 4 , T 6 do not provide correct binary information B 2 , B 4 , B 6 assigned to a code element C 1 , C 2 , C 3 .

In what follows, measures will be explained by which it is to be assured that the correct scanning elements D 1 to D 12 are employed for code word generation, i.e. those scanning elements D 1 to D 12 which respectively scan the partial areas A, B of a single code element C 1 , C 2 , C 3 .

For the selection of code elements D 1 to D 12 , which respectively scan a partial area A, B of the code elements C 1 to C 3 dependably and definitely, or of the scanning signals S 1 to S 12 generated by them, the scanning elements D 1 to D 12 , which are spaced apart from each other at the distance of the length of a partial area A, B, are compared with each other. In accordance with FIG. 4 , this comparison is performed by comparison elements T 1 to T 6 which, as a function of the difference of the incoming signals, give out an error signal F and/or a binary value B 1 to B 6 =0 or 1. The comparison results of the scanning elements D 1 to D 12 , which are arranged in a pattern corresponding to the length of a code element C 1 , C 2 , C 3 , constitute a group. Now those scanning elements D 1 to D 12 of a group are selected, whose sequence contains the fewest error signals F. In FIG. 4 the even-numbered scanning elements constitute a group, and the odd-numbered scanning elements a further group. In the relative position represented, the odd-numbered scanning elements D 1 , D 3 , D 5 , D 7 , D 9 , D 11 are selected, i.e. the output signals B 1 , B 3 , B 5 (code word CW 2 ) of the lower comparison elements T 1 , T 3 , T 5 . This described selection routine generates the second selection information A 2 that is used to control the selection of scanning signals S 1 to S 6 selected by the selection devices of FIGS. 2 and 3 . Thus, the scanning signals from a group of scanning elements, or scanning signals, are selected for further processing and for forming the measured position value POS 2 by the second selection information A 2 .

The additional information Z does not necessarily have to be only one incremental graduation, it can also include several incremental graduations. In an advantageous manner the additional information Z includes a combination of absolute codings with an incremental graduation, single- or multi-track, or of an incremental graduation with an integrated marking, such as described in DE 102 44 234 A1. In this case, in an advantageous manner, as the additional information Z the incremental graduation can also have a shorter graduation period than the length of a code element C 1 to C 5 , since the absolute coding, or the marking, is used for the determination of the definite absolute position POS 3 within a code element C 1 to C 5 .

Obtaining the first selection information A 1 from an additional information Z, which is arranged parallel with the code C, and obtaining the second selection information A 2 from the code C, has the advantage that both types of selection information A 1 , A 2 are obtained from different locations of the scale 1 . No additional hardware redundancy is required for this, because this additional information Z is simultaneously used for obtaining a measured position value POS 3 , which complements the absolute measured position value POS 1 .

A further exemplary embodiment is represented in FIG. 3 . Components with the same effect are provided with the same reference symbols in all figures. Differing from the above described explanations of the present invention, here the incremental graduation IN is only used for complementing the measured position value POS 1 from the coding C. As described above, and here schematically represented in the form of the module 11 , the second selection information A 2 is obtained directly from the code by forming the difference from scanning signals S 1 to S 12 of the partial areas A, B of the code elements C 1 to C 3 . The first selection information is also obtained from the code C, but in accordance with different criteria, namely by evaluating the edges (transitions) of partial areas A, B of the code elements C 1 to C 3 which follow each other. The module 12 for obtaining the selection signal A 1 is also only schematically indicated. Here, use is made of the fact that each code element C 1 to C 3 includes a sequence of partial areas A, B, which are designed inversely with respect to each other and therefore has an edge in the shape of a transition from a physical property to an inverse physical property. These edges are arranged in a constant pattern corresponding to the length of a code element C 1 to C 3 , so that a periodic signal can be obtained therefrom, which is interpolated in a known manner, and from this a measured position value POS 3 is obtained, which absolutely divides the length of a code element C 1 to C 3 . Reference is made in connection with this to WO 02/01160 A1, for example. The relatively roughly resolving measured position value POS 1 , and the further resolving measured position value POS 2 , as well as the finely resolving measured position value POS 3 , are linked together in the logical connecting device 7 into the resultant absolute measured position value POS 4 .

The present invention can be employed in connection with photo-electric scanning, as well as other scanning types, for example magnetic, capacitive and inductive.

The foregoing description is provided to illustrate the invention, and is not to be construed as a limitation. Numerous additions, substitutions and other changes can be made to the invention without departing from its scope as set forth in the appended claims.