GB2209101A - Optical transducer sensing - Google Patents

Abstract

A remotely detected optical transducer system has a plurality of optical transducers 68 coupled via different optical path lengths to a common signal processor and a common pulsed light source 74. Each transducer 68 produces a wavelength encoded output pulse in response to an interrogation pulse from the light source 74. The respective transducer output pulses arrive at the signal processor in a time division multiplexed form. At the signal processor the pulses are firstly wavelength de-multiplexed by a graded index rod lens and wedge grating assembly 80 which disperses the different wavelengths along a first co-ordinate direction. The simultaneous wavelength de-multiplexed output pulses are applied to the entry slit of a streak tube 86. The deflection plates 92 of the tube produce perpendicular to the slit. A ramp voltage 98 applied to the plates is synchronised with the interrogation pulses. Successive arriving groups of wavelength de-multiplexed pulses are thus separated and imaged on a phosphor screen 94. <IMAGE>

Description

OPTICAL TRANSDUCER SENSING This invention relates to the detection of signals from optical transducers.

There have been recent advances in the types of optical transducers available. A wavelength multiplexed optical position transducer is described in Optics Letters Vol 11, No.l (Fritsch and Beleim). This uses a graded-index rod lens and a glass wedgegrating assembly to disperse and recombine a broad spectrum optical signal. The dispersed light is reflected by a coded strip. The combination of wavelengths present in the recombined spectrum is a function of the position of the lens/grating assembly relative to the coded strip. The reflective areas of the coded strip may be such that the wavelengths present indicate position in a gray code, although a normal binary or another code may be employed.

In order to decode the output of the transducer, this output is dispersed again and the intensities at each wavelength are detected and thresholded to determine which wavelengths are present.

With a system such as that summarised above, it is necessary to have a single optic fibre leading to each transducer and a separate means to detect the output of each transducer i.e. a separate wavelength de-multiplexer is needed to read the output of each transducer.

One device for detecting and measuring dynamic luminous events is the streak camera. This converts time information from a luminous event to spatial information and can be used to separate very closely occurring optical events by spreading spatially the corresponding light pulses. The principles of operation of the streak camera are fully described in the IEEE Journal of Quantum Electronics, Vol. QE-20, No.12, 1984 by TSUCHIYA, Y.

According to the present invention there is provided apparatus to decode a wavelength and time division multiplexed pulsed optical signal, comprising optical input means leading to an optical wavelength de-multiplexer which disperses each pulse along a first co-ordinate direction in accordance with the wavelengths present in the pulse, and a streak tube receiving the dispersed pulses and arranged to displace the dispersed pulses in dependence upon time and in a second co-ordinate direction transverse to the first.

The de-multiplexer and streak tube accordingly disperse a succession of pulses into a two-dimensional array of luminous pulses. The first and second co-ordinate directions of this array represent wavelength and pulse number respectively.

The invention is now described in more detail by way of example with reference to the drawings in which: Figure 1 is a schematic diagram of a single optical position transducer and the means to decode its output.

Figure 2 is a diagram of an optical position transducer in its housing.

Figures 3a and 3b show schematically two arrangements for interrogating and detecting the outputs of a plurality of optical transducers.

Figure 4 shows schematically a streak camera reading the output of a wavelength de-multiplexer.

Figure 5 shows a typical output from the screen of a streak camera.

The system of Figure 1 comprises a position transducer unit 2 and a control unit 4 connected by a dual fibre optic link 6,7. The control unit has two LED's 8, the outputs of which are combined by a 2:1 coupler 10 leading to the output 12 of the control unit. The input 14 of the control unit leads to a wavelength de-multiplexer unit 16 consisting of a graded-index rod lens 18 (GRIN) and a glass wedge-grating assembly 20 typically with a groove density of 1800 lines/mm. This provides an eleven channel optical output 22.

The transducer unit 2 has an input 24 and an output 26. These are connected together by a 2:1 coupler 28. The output of this lends to a wavelength de-multiplexer 30 similar to the one at the control unit. The outputs 32 of this unit are directed by optic fibres onto a reflective coded strip 34, typically coded with gray code in which only one bit changes in adjacent code positions. This reflective strip is longitudinally movable relative to the optic fibres and can be moved in response to a displacement of the transducer shaft. The coupler 28 may be located at the control unit if there are no connectors between the units 2 and 4 or if the connector back-reflections are low, say less than -35dB. Only a single optic fibre link will then be needed.

In operation, the two LED's 8 typically launch power levels of +30dB > w into 100AKm core fibre and are coupled together by the coupler 10 to achieve an adequate power level over a sufficiently broad band. The output 12 of the control unit is connected by the optic fibre 6 to the input 24 of the transducer unit 2. The light produced by the LED's is typically a broadband spectrum, the bandwidth of which is greater than 100 nm, and is supplied in pulses whose durations are in the 10's of nano-seconds range. If a single transducer is used, the LED's may be modulated by a 10 kHz sinewave with a peak power of + 20 dB < w. With multiple transducers, typically the LED's are pulsed with a 10 ns pulses at a frequency of 1 MHz and a peak power level of + 30 dB > uw.

Light enters the transducer unit and passes through the coupler 26 to the wavelength de-multiplexer 30. The grating-lens assembly typically has a linear spectral dispersion of 75nm/mm. This gives a wavelength difference between channels of 12nm and a total wavelength difference between first and last channels of about 1iOnm if ten channels are used. The eleventh channel is used for electronic reference purposes and reads "on" in all positions of the reflective strip 34.

The dispersed light is directed through the rod lens to optic fibres 32 which in turn direct it to the reflecting strip 34. This may be made from a boro-silicate glass with a low co-efficient of thermal expansion, coated with chromium for the reflective areas.

If the light emerging from a fibre 32 strikes a reflective area, a proportion of the light is reflected back into the same fibre 32 and thence to the wavelength de-multiplexer 30 where it is recombined by the wedge-grating assembly. The recombined light then passes through the 2:1 coupler 28 and is transmitted to the output 26 of the transducer unit as a pulse containing such wavelengths as were reflected by the strip 34.

The output 26 of the transducer unit is connected by the fibre 7 to the input 14 of the control unit 4. From here the light enters the wavelength de-multiplexer 16 which acts in the same way as the de-multiplexer 30 and splits the received light into the same wavelengths at the same relative intensities that were reflected by the reflective strip.

The dispersed light is then transmitted via optic fibres to the outputs 22 where it may be directed onto an array of photo-diodes.

The'outputs of the photo-diodes may be processed in conventional manner by amplification and thresholding to yield a digital output with bit values of 0 and 1 representing the absence and presence of the various wavelengths. This digital output represents the position of the transducer strip 34 relative to the transducer demultiplexer 30.

Channels 1 to 10 provide data in gray code while channel 11 provides a continuous ON signal as a reference to regulate the LED output power via the automatic gain control of an amplifier.

The intensity of the light will be attenuated at various points in the system, the important ones being the couplers, the connections to the control unit and transducer unit, the wavelength de-multiplexers, the connections to the de-multiplexers and the reflective strip. The largest OFF signal possible may be set at -28dB and the smallest ON signal at -23dB.

The optical joints between components are effected with a fusion splicer and packaged so as to maintain the optical signal as well as possible. The losses at connectors may, however, be inconsistent from connector to connector. This may lead to variations in the amplitude of the optical signal. Other factors effecting this are, the deterioration of LED's with age, the varying degree of bending in the fibre optic cables, and varying temperatures which affect the coupler outputs. For the single transducer case described here, the variations can be compensated for electronically, by feedback control of the energising signals to the LED's 8.

Figure 2 illustrates the structure of an optical position transducer such as the one used in the above system.

A cylindrical case 36 surrounds a central slidable shaft 38, the case being typically 25mm in diameter. An end cap 40 provides a stop to limit the travel of the shaft at one end while an annular seal 42 limits it at the other end by butting against a carriage 44 fixed concentrically on to the distal end of the shaft. This carriage is supported by a roller bearing 46 and a further bearing 64 in which the shaft slides. The roller bearing is mounted opposite a read head 48 fixed through the case 36 adjacent to a glass reflective coded strip 58 mounted on the carriage.

The read head 48 is connected via eleven optic fibres 50 to a wavelength de-multiplexer 52 mounted on a bracket 53. This is connected by a further fibre 51 and a coupler 54 to input and output optic fibres 56.

A steel cover 60 attached to the side of the case 36 protects the wavelength de-multiplexer and the coupler.

The entire space within the cover is potted using silicone rubber (not shown) to protect the components against thermal shock and vibration.

The shaft 38 is free to move axially in response to a displacement force on the shaft. Typically, the end 66 is connected to an item whose displacement it is required to know by a clevis 62.

The displacement causes motion of the coded reflective strip relative to the read head 48 and thus the multiplexed output in the optic fibre 56 varies and can be read remotely at a control unit.

The ends of the optic fibres 50 are embedded in the read head 48 and are disposed along a line transverse to the axis of the shaft 38, so that each fibre 50 terminates opposite a corresponding longitudinal track on the coding strip 58. Each such track has its reflective areas disposed in accordance with the gray or other code used.

Figures 3a and 3b show two possible networks of optical transducers 68 which may be optical position transducers. In Figure 3a, the transducers 68 are connected by optic fibres 70 of differing lengths to a 1:N coupler 72. The coupler is connected to a light source 79 and a signal processor 76 by way of a 2:1 coupler corresponding to the coupler 28 of Figure 1. The source 74 can comprise elements 8 and 10 of Figure 1 while the processor can comprise elements 18, 20 and 22.

In Figure 3b, the transducers are connected by 2:1 couplers 78 to a single optic fibre fed from the light source 74 and sending signals back to the signal processor 76.

In both Figure 3a and'Figure 3b, the total optical path length from the light source to the signal processor via the transducers is different for each transducer. If the transducers are interrogated by a single pulse of light from the source 74, the responses from the various transducers will be received at different times at the signal processor. To ensure that there is a finite amount of time between the pulses arriving at the signal processor, optical delays may be included in each line. These delays are normally coils of optic fibre connected into a line to a transducer. Typical times between the responses from each transducer received at the multiplexer are of the order of 10's of nano-seconds.

The reduction in light power in passing through a 1:N coupler twice is 1/N2. This will be the loss for the circuit of Figure 3a.

For Figure 3b the power reduction will be 1/(2N)2, where N is the number of transducers coupled to the single optic fibre. A high intensity light source therefore needs to be used to produce a reasonable input to the signal processor.

The pulse responses received at the signal processor are in effect a time division multiplexed signal, namely a sequence of pulses corresponding to the transducers 68 respectively. In order to extract information from individual transducers, the signals will therefore have to be time de-multiplexed as well as wavelength demultiplexed.

Figure 4 shows a system to wavelength and time de-multiplex the outputs of a plurality of optical transducers from networks such as those illustrated in Figures 3a and 3b. A wavelength de-multiplexer 80 similar to the de-multiplexer 18, 20, 22 described with reference to Figure 1 is used to disperse the incoming pulses along a first co-ordinate direction. The output of the wavelength de-multiplexer is transmitted via a plurality of optic fibres 82 to the input 84 to a streak tube 86.

The streak tube 86 converts time information from a luminous event into spatial information. The tube combines the operating principles of a photomultiplier and an oscilloscope, i.e. the photoelectric effect, electron beam steering, and phosphorescence.

The tube comprises a photocathode 88, an accelerating mesh 90, a pair of deflection plates 92 and a phosphor screen 94. Light is applied to the photocathod 88 through the fibres, arranged in a linear array extending in the first co-ordinate direction, perpendicular to the second co-ordinate direction in which the deflection plates 92 deflect the electron beam. Figure 4 is drawn as if the deflection direction is parallel to the first co-ordinate direction but this is only because the drawing is a two-dimensional representation. The deflection direction is actually perpendicular to the first co-ordinate direction.

In the system of Figure 4, the simultaneous wavelength demultiplexed pulses resulting from each transducer output pulse are applied in parallel to the streak tube 86 via the linear array of fibres. Photons strike the photocathode 88 producing an emission of electrodes proportional to the intensity of the light. These electrons are accelerated by the accelerating mesh 90 and then pass through the deflection plates 92. A ramp voltage 98 is applied to the deflection plates to sweep the electrons across the phosphor screen 94. This ramp voltage is synchronised with the pulse repetition frequency of the interrogation pulse for the transducers.

The electrons form a luminous image on the phosphor screen. A typical screen image is shown in Figure 5 with wavelength in the vertical (first co-ordinate) direction and time in the horizontal (second co-ordinate) direction. Time corresponds to pulse number in the sequence of response pulses and hence to transducer number in the plurality of transducers ordered in accordance with their round trip delay times from light source to transducer and back to the signal processor. The outputs from each transducer appears as separate columns of streaks due to the time differences between the arrival of pulses from individual transducers. The various wavelength intensities are related to the positions measured by the individual transducers. Typically, a plurality of pulses are used to produce a single displacement signal. The combination of high pulse frequency and the time constant of the phosphor screen ensures effective integration of the results from successive pulses on the phosphor screen. The image on the screen is therefore a weighted average of the displacement over the previous group of pulses.

A multichannel image intensifier plate 96 may be included in the streak tube between the deflection plates 92 and the phosphor screen 94. This serves to amplify the electron intensity and a typical range for the electron gain is 104-106. Alternatively an externally mounted image intensifier may be used.

In order for the information from the phosphor screen to be used in a control system, some form of detector needs to be placed adjacent to it. This detector may typically be a photodiode array with some electronic processing (amplification and thresholding) to provide digital signals which can be used as the inputs to a host computer. Alternatively a vidicon tube can be used.

Thus the outputs from the transducers can be accurately wavelength and time de-multiplexed. This obviously results in a saving of components at the signal processor where, in the absence of the invention, a wavelength de-multiplexer would be needed for each transducer, and each transducer would need to be linked to the signal processor by a separate optic fibre.

The system described here uses optical position transducers.

However, the invention is not intended to be limited to position transducers, and can be used with any transducers producing a wavelength multiplexed and time-division multiplexed optical output.

Claims (19)

1. Apparatus to decode a wavelength and time division multiplexed pulsed optical signal, comprising optical input means leading to an optical wavelength de-multiplexer which disperses each pulse along a first co-ordinate direction in accordance with the wavelengths present in the pulse, and a streak tube receiving the dispersed pulses and arranged to displace the dispersed pulses in dependence upon time and in a second co-ordinate direction transverse to the first.

2. Apparatus according to claim 1 in which the wavelength and time division multiplexed pulsed optical signal comprises the wavelength multiplexed outputs of a plurality of optical transducers each coupled to a pulsed light source via a different optical path length.

3. Apparatus according to claim 1 or 2 in which the optical wavelength de-multiplexer is a glass wedge-grating assembly.

4. Apparatus according to claims 1, 2 or 3 in which the optical input means is a single optic fibre.

5. An optical transducer system comprising a plurality of optical transducers providing wavelength multiplexed outputs, all coupled via different optical path lengths to a common pulsed light source and a common signal processor, the transducers being interrogated by pulses from the light source, the response to each pulse forming a pulsed optical input signal to the signal processor, the signal processor having an optical wavelength de-multiplexer which disperses each pulse along a first co-ordinate direction in accordance with the wavelengths present in the pulse, and a streak tube receiving the dispersed pulses and arranged to displace the dispersed pulses in dependence upon time and in a second co-ordinate direction transverse to the first.

6. A system according to claim 5 in which optical delays are used with each transducer to effect time delays between the responses from the transducers at the signal processor.

7. A system according to claims 5 or 6 in which the transducers are coupled via a 1:N coupler to the signal processor where N is the number of transducers.

8. A system according to claims 5 or 6 in which the transducers are coupled via a plurality of 1:2 couplers to the signal processor.

9. A method of interrogating optical transducers comprising the steps of coupling a plurality of optical transducers with wavelength multiplexed outputs by different optical path lengths to a common pulsed light source and a common signal processor, interrogating the transducers with pulses from the light source, detecting the transducer responses to each pulse at the signal processor, optically wavelength de-multiplexing the responses by dispersing each pulse along a first co-ordinate direction in accordance with the wavelengths present in the pulse, and time division demultiplexing the dispersed pulses by displaying the dispersed pulses in dependence upon time and in a second co-ordinate direction transverse to the first.

10. A method according to claim 9 in which the time de-multiplexing is accomplished in a streak tube.

11. A method according to claim 10, characterised in that the streak tube is controlled by a ramp waveform triggered synchronously with the pulsing of the light source.

12. A method according to claims 9, 10 or 11, in which the wavelength de-multiplexing is accomplished in a glass wedge-grating assembly.

13. A method according to any of claims 9 to 12, including the step of optically delaying the outputs of the transducers to effect time delays between the responds from the transducers at the signal processor.

14. An optical position transducer comprising a coded reflective strip illuminated by a plurality of beams of light of different wavelengths from a read head, the strip being movable longitudinally relative to the read head, the beams being generated by the splitting of input white light in a dispersion unit and transmitted from the dispersion unit to the read head via a plurality of optic fibres, the beams being selectively reflected by the coded strip back into the read head and along the same optic fibres to the dispersion unit where they are recombined to form an output which is a fractional part of the input white light.

15. An optical position transducer according to claim 14 in which the reflective strip is mounted on a shaft, slidable within a cylindrical case.

16. An optical position transducer according to claims 14 or 15 in which the dispersion unit comprises a graded-index rod lens and a wedge-grating assembly.

17. An optical position transducer according to claims 14, 15 or 16, in which the dispersion unit is mounted parallel to the cylindrical case with the optic fibres bending through approximately 90D between the dispersion unit and the read head.

18. An optical position transducer according to any of claims 14 to 17, in which the dispersion unit and the plurality of optic fibres are potted in silicone rubber.

19. An optical transducer system substantially as illustrated in the accompanying drawings.