GB2106642A

GB2106642A - Position sensor

Info

Publication number: GB2106642A
Application number: GB08224170A
Authority: GB
Inventors: Brian Culshaw; Antony Harry Croucher
Original assignee: CROUCHER Ltd ANTONY H; University College London
Current assignee: CROUCHER Ltd ANTONY H; University College London
Priority date: 1981-08-21
Filing date: 1982-08-23
Publication date: 1983-04-13
Also published as: GB2106642B

Abstract

In order to achieve accurate sensing of the position of the edge of a probe or cutter shown in two extreme positions 4A and 4B sweeping out a circle 4C, use is made of fibre optics F11 and F01 and F12 and F02 arranged in opposed pairs. The probe or cutter is arranged such that its effective centre is half-way between the axes of the opposed pairs of optical fibres by equalizing the output signals on the fibres F01 and F02, and the value of this signal is then indicative of the edge position of the diameter of the swept out effective cutting circle 4C. The optical signals are preferably modulated. A lens assembly may be used to spread the beams of light traversing the gap 2. <IMAGE>

Description

SPECIFICATION Position Sensor This invention relates to position sensors and more particularly to position sensors operating to a high degree of accuracy, in the measurement of the positions of rotating cutters.

In some areas of precision engineering, such as the manufacture of fine nozzles, the position of probes and cutters or other working tools need to be determined to an accuracy of about 0.0001 inch (0.0025 mm).

Misalignment may not only lead to damage of the workpiece but also to the tool. Such accuracy has not been economically obtainable with conventional measuring techniques.

Also, with small tools high rotation speeds, e.g. up to 1000,000 r.p.m. are used. This leads to flexing of the tool so that its movement is gyratory. The use of mechanical probes to contact the tool shaft at such speeds is also made difficult by the high wear rates and the deflection that such contact might engender.

Solutions of the prior art, have provided for the location or measurement of a workpiece or a process product along a single axis, by the use of a light source and sensor arranged along a line orthogonal to both the axis and a piane containing the edge. Solutions of the prior art have provided for the location and measurement of a solid body along a single axis, by measuring the positions of the two opposite edges.

No solution of the prior art, provides for the measurement of the radius of gyration of a rotating cutter.

According to a first aspect of the present invention there is provided a method of sensing the position of the edge of a rotary cutter between two spaced sets of opposed pairs of optical fibres such that the position to be sensed will vary along an axis orthogonal to both the edge and the light path between an opposed pair; radiating an input optical signal from two fibres; detecting the illumination of the respective opposed fibres, converting an output optical signal into an electrical signal and processing the electrical signals to indicate the position the edge of the rotary cutter: the arrangement being such that the rotary cutter partially intercepts the optical signals as they traverse from the radiating spaced set to the illuminated spaced set of optical fibres.

By "spaced sets of opposed pairs of optical fibres", are meant two sets of optical fibres in which each fibre of the first set is substantially opposed to and co-axial with, a fibre of the second set. These opposed fibres are separated by a gap, which is large enough to admit the rotary cutter. The configuration is such that an optical signal radiating from the tip of a fibre of one set, will illuminate only the opposed fibre of the opposed set.

A laser, or alternatively a non-coherent light-source, provides an input optical signal for a given radiating fibre. The radiating fibres need not be in the same spaced set, but are in different opposed pairs. The same lightsource may be used for both radiating fibres of different light-sources may be employed.

The output optical signals established in the opposed optical fibres are converted into electrical signals and processed to indicate the position of the edge and radius of gyration of the rotary cutter.

Conveniently the input optical signals are provided with a distinctive feature of frequency or modulation.

By characterizing the input optical signals the effects of stray light from extraneous sources may be eliminated. Such stray light may originate from a number of sources, for example, illumination of the cutter by workshop lighting may interfere, and may vary if the shadow of an operator falls across the measuring device.

Preferably the distinctive feature of the input optical signal is a characteristic modulation of amplitude.

More preferably the input optical signal is directed through a lens assembly.

By directing the input optical signal through a lens assembly, the beam of light traversing the gap containing the cutter, is broadened.

This is especially useful if a cutter which is of greater width than an optical fibre is being examined.

More preferably, the separation between the spaced sets of optical fibres and the separation between the opposed pairs of optical fibres may be varied.

By varying the separation of the spaced sets and opposed pairs of optical fibres a range of cutter sizes may be accommodated. Furthermore, calibration and tuning of the apparatus are made easier.

According to a second aspect of the present invention there is provided an apparatus for sensing the position of the edge of a rotary cutter comprising two spaced sets of opposed pairs of optical fibres, an input optical signal source or sources for one spaced set; means for detecting the illumination of the respective opposed fibre in the opposed spaced set; means for converting an output optical signal into an electrical signal and means for processing one or more electrical signals to indicate the position of the edge or edges of the rotary cutter.

According to a third aspect of the present invention there is provided a method of sensing the position of the edge of a rotary cutter comprising placing the rotary cutter between two spaced sets of opposed pairs of optical fibres such that the position to be sensed will vary along an axis orthogonal to both the edge and the light path between an opposed pair; radiating an input optical signal from two fibres; detecting the illumination of the respective opposed fibres, converting an optical signal into an electrical signal and processing the electrical signals to indicate the position the edge of the rotary cutter and modulating the optical input signals at an integral multiple of the rotation rate frequency: the arrangement being such that the rotary cutter partially intercepts the optical signals as they traverse from the radiating to the illuminated spaced set of optical fibres.

By modulating the optical input signal at an integral multiple of the rotation rate frequency, the cutter edge positions are sensed along the same cutter diameter at each rotation of the cutter.

The invention will be further described with reference to the accompanying diagrammatic drawings, in which: Figure 1 is a sectional view showing basic elements of a fibre optic position sensor in accordance with one form of the invention; Figure 2 illustrates the outputs of the sensor at various locations of the positions to be sensed.

Figure 3 is a view similar to Fig. 1 showing the operating position of the edge of a rotary element to be sensed; Figure 4 illustrates the output of one fibre optic pair as the element rotates; Figure 5 is a block diagram of one form of electronics for the sensor illustrated; and Figure 6 is a sectional view showing the basic elements of a fibre optic position sensor in accordance with a further form of the invention.

Turning first to Fig. 1, the position sensor consists of a body 1 having a gap 2 therein and the body is adjustable along an axis 3 so that the active part of the sensor can be positioned and accurately adjusted about the position of a rotary cutter or probe 4 which is shown in Fig. 1 in an offset and stationary position. Located in the body 1 are a pair of input optical fibres Fl1 and Fl2 and opposed to them on the opposite of the gap 2 are a pair of output fibres F01 and F02.

For practical purposes, when working with a cutter or probe with a diameter d which may, for example be 0.01 inch (0.25 mm) and adapted to rotate at 100,000 r.p.m., it is convenient to have the diameters of the fibre optics should be of the order of d. The width w of the gap 2 between the opposed faces of the respective input and output fibre optics should be of the order of 2d. The limits of the sensing zones of the positions sensor are indicated by dotted lines A and B in continuation of the outer extremities of the fibre optics Fl1 and F12, F01 and F02 respectively.

As the position of the cutter or probe 4 is varied along the axis 3, and a light input is supplied to the fibres FI1 and F12, the light output on the fibres F01 and F02 varies in accordance with the curves F1 and F2 in Fig.

2. It will be seen that as the probe or cutter 4 crosses the line B, it starts to block the light entering F02 from F12 so the signal falls to a minimum at which the light path is completely blocked. As the cutter or probe 4 starts to unblock the light path to F02, it starts to block that to F01 so that F1 starts to fall as F2 starts to rise. The two curves F1 and F2 cross when the cutter or probe is at a central point. F2 rises subsequently to its normal unoccluded signal level as F1 falls to its minimum, fully occluded level. As the cutter or probe crosses the line A F1 starts to rise and reaches its maximum, unoccluded level when the cutter or probe is completely clear of the line .

If the difference between the outputs F1 and F2 is considered, then when the signals are identical on both, the rotary cutter 4 is either exactly in the centre of the space between the lines A and B of Fig. 1. The central datum point may be established whether or not the cutter or probe 4 is itself rotating Referring now to Fig. 3, when rotating, the effective position of the cutter 4, takes up extreme positions 4A and 4B sweeps out an effective cutting circle 4C with the radius of gyration of the cutter.The position of the edge of the cutter or probe, may be established by measuring the fraction of light transmitted to one of the output optical fibres Fû1 or F02.

This immediately gives the position of the edge with reference to the previously established centre. In the case of the rotating cutter, the maximum value of light transmitted is related to the position of the cutting edge at one extreme 4A or 4B of its eccentric movement, and the minimum is related to the other extreme position.

Turning now to Fig. 4, it is seen that in, for example, the fibre F01 of Fig. 3, the intensity of the output optical signal varies with the position of the cutter during one cycle of its gyratory movement.

As with the non-rotating cutter, used in the previous calibration, the output optical signal in the fibre F01 will be equal in intensity to that in the fibre F02 when the cutter is at one of its two balance positions.

This value of signal intensity may be obtained by taking the mean values of the output signals, i.e. the DC components upon which the AC variation is superimposed.

These means values, will give the average position of the cutting edge along the measurement axis 3.

The AC component of the output signal gives the deviation of the cutter from this average position. Both these measurements may be taken with a resolution and respectability of better than 0.0025 mm (0.0001 inch).

Turning now to Fig. 5. this shows a block diagram in which light emitting diodes LED1 and LED2 respectively are powered through a modulator 11 and provide a modulated input optical signals to the fibres Fl1 and F12. The signals received on fibres F01 and F02 pass to photo-detectors P1 and P2 respectively and the electrical outputs are then passed through modulation frequency filters MF1 and MF2 respectively and to precision rectifiers R1 and RS.

Signals S1 and S2, are derived from the outputs of the rectifiers by passing the signal through low pass filters LF1 and LF2 respectively. The low pass filters are set at a centre frequency which is sufficiently low to exclude the AC signals generated by the rapid, gyratory motion of the cutter. The signals passed by the filters will give the position of the edges of the cutter, which may vary comparatively slowly as the cutter moves. Substracting S1 and S2 and zeroing the difference will locate the centre of the probe cutter onto the central datum point established by the earlier balance procedure.

Signals T1 and T2, are derived from the outputs of the rectifiers by passing the signal through a filter whose centre frequency is the probe rotation speed. The signals will contain only AC component of the rectifier outputs.

The signals passed by the filters will give the deviation from the central position over a period of one rotation, and hence give the gyratory motion.

For operating over a wide range of cutter speeds, the filter may be omitted, with compromises on the noise levels of the system.

The electronic processing may be implemented in an analogue form, or may be executed digitally by sampling the signals at a point such as D1, D2 in Fig. 5. Digital processing will be in practice more flexible and will permit simple corrections to be applied for effects of component drifts etc.

The effects of component drift etc, may be completely removed by recalibrating the zero of the system with the cutter or probe out of the region AB of Fig. 1. The calibration may also be updated for exact edge location by refreshing the memory of the electronics during the entry of the cutter into the measurement zone. This calibration is referenced to the position of the measurement head in space, and is largely independent of the effects of lead screw wear, and so forth.

Fig. 6 illustrates a further form of the invention, by way of example for the location of cutters of larger diameters using a lens assembly.

The gap 5 is not of as critical a width in this version. Ideally it should be as wide as is practical, considering the iimits of mechanical stability, as the cutting lubricant, and associated "swarf" may otherwise damage or occlude the lenses.

The lenses 8 and 9, are of a focal length such that the diameter of the beams 12, is slightly over half the maximum cutter diameter to be used. The focal length is the product of the diameter of the beams and the numerical aperture of the associated optical fibre.

The fibre separation s should be greater than the beam width. The actual value of the separation is determined by the application, in particular the range of cutters to be measured.

If the minimum cutter diameter to be measured is D then a separation of approximately s = d + 0.7D has been found appropriate. The maximum cutter diameter measurable is then about s + 0.7d.

The diameter of the fibres 10 and 11 should be as large as possible, for ease of alignment. A diameter of 200-400 microns has been found suitable.

The function of this aspect of the invention is similar to that of the first aforementioned aspect. The input optical fibres 10 carry an input optical signal which radiates from the tip of the fibres 10, and illuminates the first lens assemblies 8. A parallel beam of light 1 2 propogates across the gap 5 and partially illuminates the second lens assemblies 9 which focus the optical signals on the tips of output optical fibres 11. The cutter 7 partially blocks the light beams 1 2 reducing the intensity of the output optical signal and enabling the processing means to calculate the positions of the cutter.

Various modifications may be made within the scope of the invention. For instance, under some circumstances, it may be appropriate to use more than two input point, for instance to establish the relative postions of a number of cutters, or as a technique for dealing with high resolution location of large diameter cutters. The signal processing may also take several other forms.

Similarly the use of filters and so forth will vary depending on the exact nature of the application. It will also be appreciated that although it is preferred that the input signals on fibres Fl1 and F12 should be identical this is not absolutely necessary if the parameters are adjusted to deal with related but not identical inputs.

Claims

1. A Method of sensing the position of the edge of a rotary cutter comprising placing the rotary cutter between two spaced sets of opposed pairs of optical fibres such that the position to be sensed will vary along an axis orthogonal to both the edge and the light path between an opposed pair; radiating an input optical signal from two fibres, detecting the illumination of the respective opposed fibres, converting an output optical signal into an electrical signal and processing the electrical signals to indicate the position the edge of the rotary cutter: the arrangement being such that the rotary cutter partially intercepts the optical signals as they traverse from the radiating spaced set to the illuminated spaced set of optical fibres.

2. A method as claimed in claim 1 or 2, in which the input optical signals are provided with a distinctive feature of frequency or modulation.

3. A method as claimed in claim 1 or 2, in which the distinctive feature of the input optical signal is a characteristic modulation of amplitude.

4. A method as claimed in any preceding claim in which the input and output optical signals are directed through a lens assembly.

5. A method as claimed in any preceding claim in which the separation between the spaced sets of optical fibres and the separation between the opposed pairs of optical fibres may be varied.

6. A method of sensing the position of the edge of a rotary cutter as hereinbefore described with reference to Figs. 1 to 6.

7. An apparatus for sensing the position of the edge of a rotary cutter comprising two or more spaced sets of opposed pairs of optical fibres, an input optical signal source or sources for one spaced set; means for detecting the illumination of the respective opposed fibre in the opposed spaced set; means for converting an output optical signal into an electrical signals to indicate the position of the edge or edges of the rotary cutter.

8. An apparatus as hereinbefore described with reference to Figs. 1 to 6.

9. A method of sensing the position of the edge of a rotary cutter comprising placing the rotary cutter between two spaced sets of opposed pairs of optical fibres such that the position to be sensed will vary along an axis orthogonal to both the edge and the light path between an opposed pair; radiating an input optical signal from two fibres; detecting the illumination of the respective opposed fibres, converting an output optical signal into an electrical signal and processing the electrical signals to indicate the position the edge of the rotary cutter and modulating the optical input signals at an integral multiple of the rotation rate frequency: the arrangement being such that the rotary cutter partially intercepts the optical signals as they traverse from the radiating to the illuminated spaced set of optical fibres.