GB2232544A - Position transducer - Google Patents

Description

POSITION TRIMSDUCERS This invention relates to position transducers.

Position transducers may have an optically-encoded member with one or more tracks of opaque and transparent regions and an optical read head. The transducers either give an absolute measure of position or give an incremental measure of position by measuring displacement from a known position. Where an absolute measure of position is provided, this is achieved by using several parallel tracks which are each encoded to different levels so that the most coarsely encoded track identifies absolutely a position within the first or second half of the track. The next track identifies the position more accurately to one of four quarters of the track. In this way, by using a sufficient number of tracks, the desired resolution can be obtained. This, however, can require a large number of tracks and hence a large number of read heads and their associated optical components which makes it necessary to multiplex the position data.

Incremental transducers can be made with only one track if directional information is not needed (such as with a rotating encoder confined to rotate in one direction). The track carries a simple binary clock pattern of alternate regions of different characteristics and resolution is only limited by the reading resolution of the read head. The transducer functions by counting up or down from a known starting position asthe encoded track is moved relative to the read head. Displacement in the opposite direction can be measured by using a second track to give information about the direction of displacement. Although these incremental transducers are of simple construction and can give a high resolution, they suffer from a major disadvantage in that, if there is any power loss which results in loss of position data, it is necessary to reset the transducer by returning the encoded member to a datum position. This prevents the use of such transducers in situations where safety is important and where there may be interruptions in the power supply.

W It is an object of the present invention to provide a position transducer that avoids to a substantial extent the disadvantages of previous transducers.

1 According to the present invention there is provided a position transducer comprising an encoded member and sensing means, the encoded member having a first encoded track that is movable relative to the sensing means in a direction along the track, the track having a pseudorandom binary pattern of length 2 n bits, and the sensor having store means for at least n bits, means for reading an individual one of the bits from the track into the store means such that the store means contains the last n bits read from the track, and means for determining the absolute position of the track and hence the encoded member from the contents of the store means.

The encoded member preferably has a second track parallel with the first track, the second track being encoded with bits of the same length as bits in the first track, and the sensing means including means for sensing the second track and determining the boundaries of the bits of the first track. The encoded member nay have a third track parallel with the second track, the third track being encoded with bits of the same length as the first and second tracks, the bits of the third track being displaced from those of the second track and the transducer including means sensing the second and third tracks such as to provide outputs that change at different locations along the encoded member and thereby enable the direction of movement of the encoded member to be determined. Alternatively, the sensing means may include two sensors spaced from one another along the second track such the outputs of the two sensors change at different locations along the encoded member and thereby enable the direction of movement of the encoded member to be determined.

The store means may include two stores of n bits each that are supplied with the same bits from the first encoded track, the store means being arranged such that movement of the encoded member in one direction causes one store to be filled and the other to be emptied, and movement in the opposite direction causes the other store to be filled and the one to be emptied. Alternatively, the store means may have 2n - 1 storage locations, bits from the first encoded track being supplied to the central location in the store and being displaced along the store in opposite directions according to the direction of movement of the encoded member. The encoded member may of rectangular shape, the or each track being straight and the encoded member being displaceable along a straight line parallel to the or each track. The first encoded track may be optically encoded and may have alternate optically opaque and transparent regions. Alternatively, the first encoded track may have alternate optically reflective and nonreflective regions. The sensing means may be coupled with the encoded member by time-division multiplexing.

An optical position transducer in accordance with the present invention will now be described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 is a schematic diagram of the transducer; Figure 2 shows an encoded plate of the transducer; Figures 3A illustrate changes in the transducer to 3C on displacement of the plate; Figure 4 shows a part of an alternative encoder plate; and Figure 5 is a schematic diagram of an alternative transducer.

With reference to Figures 1 and 2, the transducer includes a light source 1 which supplies optical radiation along path 2 to a beam splitter 3 of conventional construction. The beam splitter 3 supplies light to three apertures 31, 32, and 33 which align with respective straight tracks 21, 22 and 23 along an optical encoder plate 20. Light passing through the plate 20 is focussed on one end of respective optical fibres 41, 42 and 43. The fibres 41 to 43 extend to a sensing unit 50 in which light emerging from the rear end of the fibres is focussed on respective photocells 51 to 53. The photocells 51 to 53 provide electrical outputs on lines 61 to 63 to a processing unit 70 which will be described in more detail later. The processing unit 70 provides an output on line 71, representative of the position of the encoder plate 20, which is supplied to a display unit 80 or other utilisation means.

The encoder plate 20 is coupled in some movable member (not shown), such as a pressure diaphragm, the position of which is to be measured. The plate is of rectangular shape and the three tracks 21 to 23 extend in parallel with one another along the length of the plate. Each track is made up of transparent and opaque regions and the plate itself may be opaque apart from the transparent regions of the track, or transparent apart from the opaque regions of the track. The upper track 21 shown in Figure 2 has a pseudo-random binary pattern of 1 - 7 n length 2 This is also illustrated by the sequence of binary O's and I's in Figure 3A, 3B and 3C. One property of pseudo-random binary sequences of length 2 n bits is that if a window of length n bits is placed over the sequence. the position of the window along the sequence is uniquely identified by the bits appearing in the window. It is this property that is used in the present invention to enable the absolute position of the encoder plate to be determined.

For simplicity, consideration is confined to a sequence of length 2Lt, that is 16 bits, and a window of length 4 bits. In this example, the processing.unit 70 has a shift register, or similar store 72, of length 4 bits. At start up, the plate 20 is assumed to be located, as shown in Figure 3A, where the forward end of fibre 41 is aligned with an opaque region or binary 0 bit. The first location A in the store 72 is, therefore, filled with a binary 0, the other locations B to D being empty. At this stage, without any additional information, it would not be possible to identify uniquely the position of the encoder plate 20 because it could be in any of the eight positions in which an opaque region is located in front of the fibre 41. If the plate 20 now moves three bits to the left, to the position shown in Figure 3B. the obit in location A in the store is shifted via locations B and C to location D. Similarly, location C will be filled with a binary 0, location B a binary 1, and location A a binary 0. The store 72 now contains information about the forth, fifth, sixth and seventh bits along the track 21 and this now absolutely identifies the position of the encoder plate.

Because adjacent bits of the track 21 do not always differ from one another, the store 72 is clocked to -shift bits along the store in response to information derived from the other two tracks 22 and 23 which also provide information about the direction of displacement of the plate 20.

The lower two tracks 22 and 23 are simple clock tracks of alternate opaque and transparent regions, that is, alternate 0 and 1 bits. Each bit is of the same length as the bits of the upper track 21, with the track 22 adjacent the upper track being arranged in phase with the upper track so that their bits align with each other. The output from photocell 52 associated with the second track 22 provides a clock signal which, after suitable processing in a processor 73 is used to clock the shift register 72.

The output from the other photocell 53, associated with the lower track 23 is also supplied to the processor 73 and this output, in conjunctiop with that derived from the central track 22 is used to indicate the direction of movement of the encoder plate.

1 1 If, for example, the plate 20 starts in a position in which both fibres 42 and 43 are aligned with transparent regions on the respective tracks 22 and 23 then the outputs from both photocells will be 1. Movement of the plate 20 to the left will cause a change in the output of photocell 53 from 1 to 0 before the photocell 52 changes from 1 to 0. Conversely, movement of the plate 20 to the right will cause a change in the output of photocell 52 from 1 to 0 before the output of the photocell 53 changes from 1 to 0. The processor 73 monitors which photocell output changes first and, from this, derives an output indicative of the direction of movement of the plate. This output is supplied on line 74 to the shift register 72 to control the direction in which the contents of each location of the shift register are displaced.

The situation described above in relation to Figures 3A and 3B related to movement of the plate 20 to the right. Because the input to the shift register 72 is made at its left hand end, further displacement of the plate 20 to the right would simply result in the contents of location D overflowing and a new bit being entered into location A, so that the shift register maintains data on four adjacent regions of the track 21. However, if the plate 20 was to be displaced from the position shown in Figure 3B to the left by one bit, to the position shown in Figure 3C, this would result in the contents of the stores A to D being transferred one location to the left with the original contents of store A overflowing to the left and with store D is being emptied. In this state, the shift register 72 only contains data on three successive regions of the track 21, namely 100. In general, three bits will not be sufficient absolutely to identify the location of the plate 20. For example, as can be seen from Figure 3, there are two locations along the plate where a three bit window would enclose the bits 010, or 011. By itself, therefore, the upper track would not be sufficient to absolutely define the position of the plate if it is free to move in both directions. However, in conjunction with the output from the processor 73 its position can be uniquely identified regardless of direction. This is because the processor can determine from tracks 22 and 23 that the plate 20 has been displaced by one bit to the left from the position 0100 so that its position can be absolutely defined.

If power should be lost to this system, so that the contents of the shift register 72 are lost, this will result in location A being filled again immediately after power is restored. Although this will not uniquely identify the position of the encoder plate 20 immediately, as soon as the plate has been displaced three bits to the right, this will again cause the shift register to be filled and will enable all subsequent positions, following displacement in either direction, to be identified. With t.

1 this system, it will be necessary to ensure that the plate is displaced to the right following loss of position information, but this is generally less than would be necessary with a purely incremental encoder.

In a modification of the invention it would be possible to employ two shift registers supplied with the same output from the photocell 51 but from different ends of the shift register. In this way, one shift register would be filled by displacement of the encoder plate to the left while the other one is filled by displacement to the right. Following power loss with such an arrangement, displacement of the plate 20 by three bits in either direction would provide the shift register with enough information to identify absolutely the plate position. Alternatively, a single shift register of seven (2n - 1) storage locations could be used which is supplied at a central location, this is, its forth location, with data from track 21.

In some transducers, the encoder plate may only move in one direction. For example, the plate may be a disc rotated in one direction about its centre and with encoded tracks of circular shape concentric of the disc. Alternatively, the plate may be a looped, endless belt that rotates in one direction and on which the tracks extend along the belt. With such transducers it is not necessary to have any direction information and only one shift register is necessary.

Where direction information is needed, this need not be provided by two separate tracks out of phase with one another. Instead, as shown in Figure 4, a single track 221 could be used with two receiving fibres 421 and 431 spaced apart along the track, out of phase with one another, so that they change in state at different times.

The transducer encoder plate need not have transparent and opaque regions. Instead it could be encoded by tracks of reflective and non-reflective regions, as shown in Figure 5. In this arrangement a time-division multiplexing arrangement is employed. A source 100 generates pulses of lightwhich are supplied via a first coupler 101 and a fibre-optic cable 102 to a second coupler 103 with three outputs in the form of delay lines 104, 105 and 106. The delay lines introduce different delays into the pulses which are supplied to the read head 107 which in turn focuses the light pulses onto three respective reflective tracks of an encoder plate 108. Light pulses reflected back from the plate 108 return via the respective delay lines 104 to 106 to the second coupler 103 and from there, via the single cable 102, to the first coupler 101. A sample of these reflected pulses is supplied via an optical guide 110 to a processing unit 111. The tracks on the plate 108 are identified from the different delays introduced into the 1 - 13 path length travelled by the light pulses supplied to and reflected from the different tracks. Similar time-division multiplexing arrangements could be used with other form of encoding.

It will be appreciated that the invention is not confined to use with optically encoded members but could be used with any other form of encoding, such as magnetic encoding or by use of electrically conductive and non-conductive regions.

14 -

Claims (15)

CLAIMS 1.

1 A position transducer comprising an encoded member and sensing means, wherein the encoded member has a first encoded track that is movable relative to the sensing means in a direction along the track, the track has a pseudo-random binary pattern of length 2n bits, and wherein the sensing means has store means for at least n bits, means for reading an individual one of the bits from the track into said store means such that the store means contains the last n bits read from the track, and means for determining the absolute position of the track and hence the encoded member from the contents of the store means.

A transducer according to Claim 1, wherein the encoded member has a second track parallel with the first track, wherein the second track is encoded with bits of the same length as bits in the first track, and wherein the sensing means includes means for sensing the second track and determining the boundaries of the bits of the first track.

A transducer according to Claim 2, wherein the encoded member has a third track parallel with the second track, wherein the third track is encoded with bits of the same length as the first and second tracks, wherein the bits of the third track are displaced from those of the second track, and wherein the transducer includes means sensing the second and third track such as to provide outputs that change at different locations along the encoded member and thereby enable the direction of movement of the encoded member to be determined.

A transducer according to Claim 2, wherein the sensing means includes two sensors spaced from one another along the second track such that the outputs of the two sensors change at different locations along the encoded member and thereby enable the direction of movement of the encoded member to be determined.

A transducer according to any one of the.preceding claims,' wherein the store means includes two stores of n bits each that are supplied with the same bits from the first encoded track, and wherein the store means are arranged such that movement of the encoded member in one direction causes one store to be filled and the other to be emptied..and movement in the opposite direction causes the other store to be filled and the one to be emptied.

6.

z A transducer according to any one of Claims 1 to 4, wherein the store means is a store with 2n - 1 storage locations, wherein bits from the first encoded track are supplied to the central location in the store and are displaced along the store in opposite directions according to the direction of movement of the encoded member.

A transducer according to any one of the preceding claims, wherein the encoded member is of rectangular shape, wherein the or each track is straight, and wherein the encoded member is displaceable along a straight line parallel to the or each track.

1 17 - A transducer according to any one of the preceding claims, wherein the said first encoded track is optically encoded.

A transducer according to Claim 8, wherein the said first encoded track has alternate optically opaque and transparent regions.

10.

11.

12.

13.

A transducer according to Claim 8, wherein the said first encoded track has alternate optically reflective and non-reflective regions.

A transducer according to any one of the preceding claims, wherein the sensing means is coupled with the encoded member by time-division nultiplexing.

A transducer substantially as hereinbefore described with reference to Figures 1 to 3 of the accompanying drawings.

A transducer substantially as hereinbefore described with reference to Figures 1 to 3 as modified by Figure 4 of the accompanying drawings.

18 -

14.

A transducer substantially as hereinbefore described with reference to Figures 1 to 3 as modified by Figure 5 of the accompanying drawings.

1

15.

Any novel feature or combination of features as hereinbefore described.

Published 1990atThe Patent Office,State House, 6671 High Holborn. London WClR4TP. Further copies maybe obtainedfroM The Patent Office. Sales Branch, St Mary Cray, Orpington, Kent BR5 3RD. Printed by Multiplex techniques ltd. St Marl Cray, Kent, Con. 1/87