OPTICAL SENSING
The present invention relates to optical sensing methods and arrangements.
The invention relates particularly but not exclusively to methods and arrangements for determining distance (including multiple distances to a scanned surface to determine the surface co-ordinates), velocity and acceleration. However other parameters such as refractive index for example or parameters convertible to distance, velocity, acceleration or refractive index can also be measured by the arrangements and methods of the invention.
In one aspect the invention provides a method of measuring a parameter comprising the steps of transmitting a signal through two arms of an interferometer, the signal being modulated or pulsed at a known frequency, varying the relative effective path length of the arms in dependence upon the parameter, varying said frequency to obtain at least two cross-correlation peaks between interfering components of the signal from the respective interferometer arms, determining the frequency difference between the cross-correlation peaks and generating an output signal indicative of the parameter in dependence upon the frequency difference.
In a related aspect the invention provides apparatus for measuring a parameter, the apparatus comprising an interferometer, means for transmitting a modulated or pulsed signal through two arms of the interferometer, means for varying the relative effective path length of the arms in dependence upon the parameter, means for varying the modulation or pulse repetition frequency of the signal to obtain at least two cross-correlation peaks between interfering components of the signal from the
respective interferometer arms, and output means responsive to the frequency difference between the cross-correlation peaks to generate an output signal indicative of the parameter.
Preferably the signal is an optical signal. However, in principle the interferometer could be e.g. a microwave interferometer.
The above apparatus is suitable for absolute position measurements over ranges of a few tens of millimetres to a few metres with high resolution and using a low power signal.
The apparatus can optionally include a transducer arranged to vary the position of a reflective element or the refractive index of a light-transmissive element in one arm of the interferometer to measure a parameter to which the transducer is sensitive.
In one embodiment the apparatus comprises signal gating means in one or both arms of the interferometer and means for applying a switching or modulating signal to the signal gating means. Such signal gating means may allow operation with only low levels of returned power from the target reflector.
Preferably the amplitude or modulation depth of the signal is controlled by e.g. an AGC circuit.
In a further aspect the invention provides scanning apparatus comprising means for generating a laser beam in a first laser cavity, beam-directing means arranged to direct the laser beam at an object to be scanned, means for measuring a frequency of
at least one external cavity lasing mode or the frequency interval between such lasing modes associated with an external cavity defined by an exit face of the first laser cavity and the point of incidence of the laser beam on the object to determine the distance to the point of incidence, means for scanning the beam over the surface of the object, and processing means arranged to correlate the instantaneous scanning position and distance to derive the co-ordinates of the scanned surface.
Preferably the means for generating the laser beam is a laser diode and the measuring means is arranged to measure the frequency of an A.C. electrical signal in the laser diode, the signal being associated with said external cavity lasing mode.
Preferably the scanning means is arranged to scan the beam in two orthogonal scanning directions.
In a related aspect the invention provides a method of determining the co-ordinates of a scanned surface of an object, the method comprising the steps of scanning the surface with a laser beam using the above-defined apparatus and measuring a frequency of at least one external cavity lasing mode with a laser beam using apparatus as defined above and measuring a frequency of at least one external cavity lasing mode or frequency interval between such lasing modes associated with an external cavity defined by an exit face of the first laser cavity and the point of incidence of the laser beam on the object, the surface of the object being a diffuse reflector of the laser beam.
In another aspect the invention provides scanning apparatus comprising a plurality of laser diodes each arranged to direct a laser beam at an object to be scanned, and
switching means arranged to switch respective alternating electrical signals from the laser diodes to a common input of a signal processor in serial fashion, the signal processor being arranged to determine the frequency of at least one external cavity lasing mode or the frequency interval between such lasing modes of each laser in an external laser cavity defined by an exit face of the selected laser diode and the point of incidence of the laser beam on the object and hence to determine the distance to the point of incidence or a function of said distance.
Preferred features are defined in the dependent claims.
Preferred embodiments are described below by way of example only with reference to Figures 1 to 20 of the accompanying drawings, wherein:
Figure 1 is a schematic block diagram showing a laser arrangement for measuring thickness;
Figure 2 is a plot of output power : lens position in the above arrangement;
Figure 3 is a further plot of output power : lens position in the above arrangement;
Figure 4 shows a variant of the arrangement shown in Figure 1 employing a transversely movable wedge and a diffusely reflective surface;
Figure 5 shows an arrangement for measuring velocity;
Figure 6 shows one initial signal processing arrangement for the embodiment of Figure 5;
Figure 7 shows a plot of speed : time obtained from the embodiment of Figures 5 and 6;
Figure 8 shows a similar plot obtained from a commercially available position sensor;
Figure 9 shows scanning apparatus in accordance with the last - mentioned "scanning apparatus" aspect of the invention;
Figure 10 shows scanning apparatus in accordance with the other "scanning apparatus" aspect of the invention;
Figure 1 1 shows a variant of the embodiment of Figure 10;
Figure 12 is a diagrammatic representation of interferometer apparatus in accordance with the "interferometer" aspect of the invention;
Figure 13 shows schematic plots of pulse repetition frequency : time and intensity : time in respect of the optical signals in the apparatus of Figure 12;
Figure 14 is a schematic block diagram showing the RF drive arrangement used in the embodiment of Figure 12;
Figure 15 is a spectrum of the cross-correlation peak obtained by the embodiment of Figures 12 to 14, but omitting the AGC block from Figure 14;
Figure 16 is a schematic block diagram of an external cavity spectral displacement sensor;
Figure 17 is a plot of peak wavelength : lens/reflector position for the sensor of Figure 16;
Figure 18 is a schematic block diagram showing a laser arrangement with an integral sample cell for level, pressure, chemical (e.g. pollution) concentration sensoring or detection;
Figure 19 shows a variant of the arrangement of Figure 18;
Figure 20 shows a further variant of the arrangement of Figure 18;
Figure 21 shows a frequency measurement arrangement for use in the above embodiments;
Figure 22 shows a variant of the above frequency measurement arrangement, and
Figure 23 shows a further frequency measurement arrangement for use in the above embodiments.
Referring to Figure 1 , a laser diode is biased at constant current by a power supply 34 and generates a laser beam 4 which is collimated by a lens 2 adjacent the right hand laser facet (preferably anti-reflection (AR) coated, although this is not essential) of the laser diode.
A lens 2A adjacent the opposite exit face may be use to focus the beam onto a photodetector 5.
A further lens 28 movable in the longitudinal direction as indicated by arrow A focuses the beam on the front or rear face of a light-transmissive sample 20. In general, both faces of sample 20 will reflect some light, as indicated at 4.
The rear facet of the laser diode is monitored by photodetector 5, and a large increase in optical output power occurs when the position of focus falls on either facet of the sample 20 under test. The thickness t of the sample can then be found from the simple expression: t = dn[l + ] (1.1) where d is the distance through which the lens has been scanned, while n is the refractive index of the material of the sample under test, α is a small dimensionless constant which can be obtained from standard geometrical optics and is dependent upon the physical properties of the optical system used. Therefore measurement of the focusing lens position provides an accurate measure of the thickness of the object under test.
By way of example Figure 2 shows the results of a measurement of intensity scan performed upon a glass microscope slide. It can be seen that two sharp peaks arise
due to reflections from the front and rear facets of the slide, which occur 0.71 mm apart. When the refractive index of glass is taken into account, it is found that the thickness of the glass slide is l.Oό±O.O l mm. This is a result which has an accuracy greater than that of a standard micrometer gauge.
Again by way of an example and to illustrate the potential resolution of the system, a microscope cover slip was also examined. This was found to have a thickness of 186±3 μm as shown in Figure 3. The FWHM of the reflection peaks is of the order of 25 μm suggesting that thicknesses <50 μm could easily be resolved. This figure could be reduced by careful choice of coated optics.
In the embodiment of Figure 4 the optical path length between the lens 2 and the device 20 under test is varied by an optical wedge 100 or other similar apparatus. The wedge is moved transversely to the optical axis as indicated by arrow A and in so doing changes the optical path length. An alternative to this system would be to use a rotating prism to produce the same effect. However any suitable optical system which can vary the position of the focus of the lens may be used.
Object 20 could be a diffusely reflective opaque object rather than a transparent object as in Figure 1.
The above embodiments have the advantage of non-contact sensing and are applicable to liquids as well as solids. They can also be used in level detection applications as well as medical applications or hazardous environment applications.
There is a demand in many applications, (particularly in aerospace) for simple low cost optical sensors to measure speed and acceleration.
Such a system may be developed by adaptation of the system used for position measurement using external cavity laser feedback. This may take the form of repeatedly differentiating the output of this sensor system with respect to time. This could be done using either analogue or digital signal processing techniques.
By way of example, the experimental system used in Figures 5 and 6 was used to perform an off-line velocity measurement demonstration.
Figure 5 shows a laser diode whose optical output is collimated by a lens 2. In this embodiment the collimated optical beam then passes through a beamsplitter 1 10 which splits off some of the optical signal to a photodetector 5 via a focusing lens 2A, for alignment purposes. The remainder of the light has its path folded by a corner cube retroreflector 10, before being reflected by a mirror 120. This mirror forms the external reflector of the external cavity with the laser diode.
In this embodiment the optical sensor is calibrated as it would be for use in normal displacement sensor applications. A characteristic response of voltage output as a function of position over its full range of operation is obtained prior to use. This is then used to calibrate the sensor's response for an off line velocity measurement.
The external reflector 10 of the optical sensor was subjected to a series of preprogrammed movements of constant velocity, in order to test the sensor's suitability for velocity measurements. Output data from the sensor was then logged by
computer to be examined off-line. Simple digital processing techniques were then employed in order to provide a measure of velocity as a function of time.
This embodiment was tested for velocity measurement capability. A direct velocity calibration to the system was provided by comparison with processed data obtained from an RDP Electronics Ltd LVDT unit. The velocity data from the optical sensor is presented in Figure 7 whilst the data obtained from the RDP transducer is shown in Figure 8.
Examination of the above data shows that a reasonable measure of velocity was obtained, which was accurate in the worst case to less than 10% of the measured value. However this figure can be greatly improved with the use of more powerful and sophisticated processing algorithms and is shown only to give proof of principle of operation.
The principle which allows the velocity to be measured can be extended in order to make a measurement of acceleration.
Figure 6 shows a laser diode arrangement comprising circuitry 13 suitable for use with the diode laser 1 of any of the preceding embodiments. Laser 1 is connected to a DC and RF port of a bias T network 33 which feeds a DC bias to the laser from a suitable constant current source 34 (connected to its DC port) and outputs an RF signal from laser 1 via its RF port to circuitry comprising RF amplifier 45, bandpass filter 35 and a further RF amplifier 36.
An automatic gain control (AGC) system (37 to 40) is utilised, if required, to maintain the power of the RF at the bandpass filter 35 at a constant level. The output from the bandpass filter is then filtered by a high pass 'shaping' filter 41. This may be any filter whose transfer function varies as a function of frequency over the passband of the bandpass filter. An example of this is a simple high pass filter whose transition region coincides with the passband of the bandpass filter.
This it can be seen that the amplitude of the RF output of the shaping filter is dependent upon the frequency of the mode in question. Therefore an RF power meter or other suitable detector device 42, such as for example, a RF Schottky detector diode is used to provide a measure of the frequency of the particular external cavity mode. This feeds a DC amplifier and a data acquisition card 44 in a personal computer (not shown) .
The filter decoding method has been demonstrated to be capable of achieving resolutions of less than 1 micrometer. However, with more specialised filter systems, specifically designed for small range / high resolution operation, this figure can be substantially improved. Therefore it is envisaged that certain embodiments may provide sub-micron accuracy in the future.
Another embodiment (Figure 21) employs a phase locked-loop based system. In this embodiment, a phase-locked loop comprising a phase-sensitive detector 150 coupled to an input of a controller 147 is used to lock a local oscillator 148 to the peak frequency of the input signal. The frequency of this local oscillator is then measured using a frequency meter 146 which has an input connected to an output of a splitter.
Again the system has the potential of being single or multi-chanelled, with a bandpass filter 135 used to select the particular mode of interest.
In this decoding embodiment, the measurement is made upon a clean local oscillator signal, and therefore this system possesses inherently better noise immunity than the simple frequency counter system. Thus, once the phase-locked loop has achieved its initial lock, it should be able to track the position of the external cavity mode as long as it stays within the passband of the bandpass filter. If the lock is held, then an extremely accurate measurement may be made since the frequency measurement is made upon the extremely low noise local oscillator signal which originated from the laser.
The embodiment of Figure 22 utilises hetorodyne decoding. An amplifier 145 feeds the signal to an image-rejection bandpass filter 135' and thence to an input of a mixer 151 whose outer input is coupled to the output of a local oscillator 148. The frequency of local oscillator 148 is varied in a sawtooth fashion by a ramp generator 155.
Low pass filter 135' blocks any unwanted image frequencies which may degrade the heterodyne process. In the most simple embodiment of the heterodyne system the local oscillator is swept periodically in a sawtooth fashion and its output mixed with the input signal. As the input to the local oscillator is swept through its range of frequencies, different input frequencies are successively mixed to pass through an IF amplifier 152 and bandpass filter 153. The signal is then detected by a detector 154. Therefore the frequency spectrum can be output as a time domain waveform. With the use of simple timing circuitry (not shown) which correlates the instantaneous
frequency of local oscillator 148 with the instantaneous amplitude of the signal detected by detector 154, a measurement can then be made of the frequency separation of the external cavity modes, thus providing position information.
In some applications the use of timing circuitry may not provide a precise enough measurement. In this case a FFT based system could be used. In this variant, the output from the heterodyne system would be input to an A/ D converter and acquired by a computer for digital signals processing (DSP). An FFT or other suitable algorithm could be performed which would be used to assess the repetition rate of the time domain signal. This is in reality the frequency separation of the time domain signal. The resolution of such a system may be enhanced using standard DSP techniques such as zero padding. To date this embodiment has been used in an off-line situation and has produced results accurate to better than 0.2% of full measurement range.
One further embodiment of the heterodyne decoding principle is shown in Figure 23 and is arranged to perform heterodyne peak tracking. In this embodiment, the input signal is subjected to initial signal processing (if required), by an automatic gain controlled amplification block 156 and is then bandpass filtered by a filter 135' in order to produce a signal which consists of only one external cavity mode which remains at a constant power at the input to a mixer 151 of the heterodyne stage. Heterodyne detection is then carried out in such a way that the mixer output IF frequency corresponding to the external cavity peak remains at the IF output of the mixer. In order to do this, the amplitude of the amplified IF output signal from IF amplifier 152 is analysed and connected in a feedback loop comprising an IF bandpass filter 153, power detector 154 and peak detection controller 157 which controlls the frequency of a local voltage controlled oscillator (VCO) 148. The peak
detection controller 157 senses changes in the detected power and changes the control voltage of the local oscillator in such a manner as to move its output frequency towards the peak of the filtered RF input signal. The frequency of the voltage controlled oscillator is then measured using a known frequency meter arrangement providing a stable, digital output.
Multiplexing of the external cavity optical displacement sensor can be achieved by using a multilaser approach. This can provide the advantage of a large number of sensor heads, which share the same signal processing circuitry, to reduce component and operating costs. One proposed system architecture is shown in Figure 9.
Many sensor head subsystems consisting of the sensor mechanics, laser, bias current and bias T's could be multiplexed using such a system. The RF output from each sensor is routed via a power combiner 140 to a single unit 150 containing all the RF signal processing components necessary for correct decoding of the sensor signal. (This is provided that only one sensor head needs to be addressed at any one time - if more than one is required to be addressed at any time, then more signal processing units are needed.) Once the RF signal has been reduced to a low frequency analogue or digital signal (block 160), the calibration data for the particular laser in operation is taken into account so that an accurate output reading is then achieved (at output block 180). Block 160 receives a calibration data from block 180 for each laser subsystem as it is switched in sequence to combiner 140.
An alternative multilaser system application would be to arrange a set of non-contact sensor heads in a two dimensional configuration, in order to perform an imaging type measurement. This could involve a multiplexed architecture as described above or in
the case when contact sensor heads need to be accessed simultaneously, the normal discrete architecture used in the case of a single sensor, would be used in parallel.
Such a multisensor application would have potential applications in the fields of imaging or medical sensing as well as general industrial sensing applications.
The laser sensor subsystems of Figure 9 could be arranged in a linear to two- dimensional array and directed onto respective adjacent portions of a surface whose co-ordinates are to be determined.
Figure 10 shows an alternative embodiment in which a beam 4A from a single laser diode 1 is scanned vertically by a mirror 200 which is tiltable about a horizontal axis XI as shown. The beam 44 from mirror 200 is then scanned horizontally by a mirror 210 which is tiltable about an axis X2 which axis is inclined at 45° to beam 4A and lies in the (vertical) plane of beams 4A and 4B. The resulting beam can then be positioned as required, e.g. it can be scanned horizontally and vertically in rastor fashion over object OB. A processor PR determines the frequency of an external cavity lasing mode and correlates this with the instantaneous scanning position of the beam.
Figure 11 shows an alternative scanning arrangement in which a lens 220 is movable transversely as indicated by arrow A to scan the beam 4 over object 20.
The RF drive arrangement as shown in Figure 14 can be used to energise the laser diode 1.
Referring to the Michelson interferometer arrangement shown in Figure 12, a simple low cost semiconductor laser diode 1 is used to send short optical pulses (of around 30 to 50 picoseconds duration) or a high speed modulated optical signal, to the target 20 which is incorporated in one arm of the interferometer. The other arm of the interferometer comprises a static mirror 20'. Target 20 reflects the pulses or modulated signal to a receiver 5 via a beamsplitter 110. Thus the two beams are initially split and subsequently combined after reflection by the beamsplitter 100. Target 20 can be moved by a transducer TR which is sensitive to an external parameter and moves target 20 independence upon the parameter.
Unlike time of flight range-finding systems, rather than using a time delay technique to determine optical path length which would require the use of extremely high speed detectors and electronics, the pulses are detected using a novel optical cross correlation technique. Also, for small to intermediate ranges, time of flight measurements are notoriously inaccurate due to the limitations placed upon the system by the maximum speed of operation of the high frequency electronics required for timing.
This technique involves the use of the Michelson interferometer to achieve cross correlation between two optical pulses or between two periods of an RF modulation.
By modulating both arms of the interferometer, e.g. using modulators 190 having modulation frequencies of fi and f2, a measurement can be made free from any background interference, thus allowing successful operation using low optical powers with a much higher sensitivity than conventional range-finding sensors. If this
technique is required, then the modulation can be carried out using liquid crystal modulators or other suitable apparatus.
To achieve accurate position sensing, the laser pulse repetition or modulation rate is linearly swept over a range of frequencies so that at least two cross correlation peaks are produced. Then rather than measuring the profiles in detail, the cross correlations are merely peak detected as depicted in Figure 13. The frequencies at which the two peaks occurred can then be obtained by considering the frequency sweep rate, so that the following relationship can be obtained: c d
where c is the speed of light an /A and fβ are the frequencies at which the cross correlation peaks A and B occur respectively. The quantity d is the optical path difference between the zero path difference position of the interferometer and the target.
The system shown in Figure 14 was used to drive the RF signal to the laser diode. In order to keep the input RF power level as constant as possible, an automatic gain control (AGC) system 37 is employed in this embodiment. This counteracts any variation in RF signal strength seen by the laser due to impedance mismatching and sweep oscillator output power variations. RF sweep oscillator 39 feeds gain /attenuation block 30 which feeds the RF port of a bias T block 33 via AGC block 37.
A D.C. constant current power supply feeds the DC port of bias T block 33 and the laser diode 1 is connected to the DC/RF port of the bias T block.
An example of the output of a variant of such an embodiment described above is given in Figure 15. Whilst the target reflector position remained constant the RF frequency was swept from 1 GHz to 2 GHz. If the path difference of the interferometer is greater than 7.5 cm a cross correlation peak should be observed. However, in this example the AGC system was not used and as can be seen in Figure 15 variations in the fringe depth can be seen. These are due to the variation in RF power input to the laser and impedance mismatches.
An external cavity laser system can be used to measure small movements by monitoring the wavelength of the light emitted by the laser. The range of such a sensor would be of the order of 100 microns and would be designed to have a high resolution.
Referring to Figure 16, in this embodiment a collimated beam 4 from the laser 1 and lens 2 is focused onto an external reflector 20 by a lens 2A which if properly aligned forms an external cavity with the laser. For optimum results an anti-reflection (AR) coated laser is used. Use of an AR coated laser facet allows an increased optical feedback to the laser and hence an increase in the measurable effect.
The laser is driven continuous wave (cw) with a dc bias current. As the relative position of the lens with respect to the external reflector is varied as shown by arrow A the optical spectrum is monitored using an optical spectrum analyser or monochromator or other suitable chromatic measurement apparatus.
It is found that the peak wavelength of the optical spectra varies as a function of the relative position of the external reflector constituting the external cavity.
By way of example, Figure 17 shows some typical results from the system depicted in Figure 16. Here the peak wavelength of the laser output spectrum has been plotted as a function of position. It can be seen that a variation in wavelength range of approximately 55 nm was achieved for a change in lens position of 40 microns. These results suggest that the potential resolution of this system would be less than 1 micron.
This embodiment would be useful for applications where a measurement of sub- micron accuracy are required. It also may have applications in the area of medical sensing as well as other specialised industrial applications.
The concept of the external cavity optical displacement sensor may be applied to the field of chemical and environmental sensing. Here it may be applied in many different ways such as, for example, level sensing in tanks and vessels, contamination and pollution sensing and pressure sensing.
By way of example, Figure 18 shows an arrangement where a chemical cell 46 has been introduced into the external cavity of the sensor. Now, if the content of the chemical cell changes in some way, in which its refractive index is affected, this will introduce a variation in optical path length within the cavity. If the original refractive index was m, and after some change becomes n∑,, then the system sees an effective change in optical path length Δ/ of;
where t is the thickness of the contents of the chemical cell. Therefore this change will give rise to a change in the external cavity RF modes generated by the system, which can then be measured. The change in the frequency of a particular mode may be given approximately by:
Δ/ = . Al, (7.2)
2P
where I is the total optical cavity length and c the speed of light in vacuum. Laser 1 is coupled to the DC/RF port of a bias T network 33, whose DC port is coupled to a constant current power supply 34 and whose RF port is coupled via RF amplification block 36 to signal processing electronics 45, which may be similar to blocks 35 to 44 of Figure 6 for example.
As a further example, Figure 19 shows a further possible embodiment where an arrangement is employed to measure path length changes. This embodiment uses a method analogous to that used for thickness measurements (see the embodiment of Figures 1 and 4) and employs a longitudinally movable focusing lens 2B as indicated by arrow A.
In this embodiment the lens would track any changes in optical path length by locking to the peak in optical power due to reflection from the mirror. This would require the lens to move accordingly so that the position of focus of the lens occurs on the mirror facet. Measurement of the change in lens position will therefore provide a measure of the change in optical path length.
By way of another example, a further possible embodiment is depicted in Figure 20. Here the mirror in Figure 19 has been dispensed with and instead, the lens 2A will track reflections from the far side of the chemical cell. The parts of Figures 19 and 20 which are similar to the correspondingly numbered parts of the other embodiments will not be described further.
In the previous sections many examples of potential sensor designs have been introduced which depend upon the principle of reflection from an external reflector which may be the device under test, or alternatively may be mechanically connected to it. In most of the embodiments described previously a plane mirror or similar apparatus has been used to provide sufficient reflective feedback to the laser to allow correct operation of the sensors. However, in most cases sufficient feedback may be obtained by using a diffuse reflector instead. This may open a whole new set of sensor applications to the main technique of external cavity feedback. However, since to observe the smallest possible effect only extremely a small reflection is required in many applications the device under test may be examined directly without coupling any reflector to it.
Therefore the use of diffuse reflectors may allow further applications in the line of medical sensing, environmental sensing and pollution monitoring.
Also in many of these cases, changes in the signal to noise ratio and frequency shifts of the external cavity RF modes may be used to sense changes in optical cavity loss. This may be due to increased pollutant levels for example.