Multi-channel array processor

This application claims priority from U.S. Provisional Patent Application Ser. No. 60/637,959, filed Dec. 21, 2004.

The present invention is generally related to fiber optic array signal processors, and more particularly to an improvement over linear array signal processor systems that interrogate Fabry-Perot sensors.

The linear array signal processor (LASP) system as seen in FIG. 1 interrogates Fabry-Perot sensors 10 that have gaps 12 ranging from 5 to 25 μm, a white light tungsten lamp 14 with spectral intensity as shown in FIG. 5 . Light 15 is delivered to the sensor 10 through a 2×1 coupler. The light 15 is modulated by the sensor 10 as shown in FIG. 2 and is reflected off a cylindrical mirror 16 through a Fizeau wedge 18 , with an optical thickness that ranges from 5 to 25 μm, and onto a linear silicon CCD array 20 with many pixels.

The LASP system can be multiplexed with many channels sharing a single microprocessor, however, each channel must be interrogated in time by switching each channel on and off. With the existing system, any changes that occur to the environmental parameter associated with any channel while another channel is being interrogated is lost. When many channels are multiplexed in time with the existing system, the update rate per channel is slow. The present invention provides for a system that is able to process many channels simultaneously and overcomes the limitations of the existing system.

Objects and advantages together with the operation of the invention may be better understood by reference to the following detailed description taken in connection with the following illustrations, wherein:

FIG. 1 is a schematic representation of the LASP system of the present invention;

FIG. 2 is a graphical representation of white light modulated by the Fabry-Perot sensor;

FIG. 3 is a graphical representation of the correlation burst from the raw signal on the CCD array;

FIG. 4 is a graphical representation of an LASP signal filtered to normalize the signal;

FIG. 5A is a graphical representation of the spectral intensity of Tungsten Quartz Halogen lamps used in the LASP system of the present invention; and

FIG. 5B is a graphical representation of the spectral sensitivity of Silicon Photodiode;

FIG. 6 is a functional block diagram of an electronic circuit of the present invention;

FIG. 7 is a graphical representation of the sensor gap versus pixel number;

FIG. 8 is a diagrammatical representation of a multi-channel LASP of the present invention; and

FIG. 9 is a diagrammatical representation showing the location of one optical fiber in relation to 2-D CCD and the mirror.

FIG. 10 Diagram showing location of one fiber in relation to 2-D CCD and lens.

While the present invention is described with reference to the preferred embodiment, it should be clear that the present invention should not be limited to this embodiment. Therefore, the description of the preferred embodiment herein is illustrative of the present invention and should not limit the scope of the invention as claimed.

Reference will now be made in detail to the preferred embodiment of the invention as illustrated in the accompanying drawings. The present invention discloses a method for multiplexing many fiber optic channels into one LASP signal conditioner.

Within the signal processor 30 , the modulated light 31 from the sensor exits the fiber 32 as a cone shaped beam 34 and is redirected from the mirror 36 as a line with a Gaussian distribution. When the light passes through the Fizeau wedge cross correlator 38 , a correlation burst as shown in FIG. 3 is created and detected by the CCD array 40 . The correlation burst occurs at the precise pixel location along the CCD array 40 where the optical thickness of the Fizeau wedge 38 is precisely the same as the optical path length of the Fabry-Perot gap.

In order to determine the peak intensity given the characteristic Gaussian distribution, normalizing the readings based on the distribution resulting in a signal that looks like the signal shown in FIG. 4 is required. A second signal processing algorithm is used to determine the precise pixel or location of the correlation burst along the CCD array. The signal conditioner then converts the Fabry Perot gap measurement into the appropriate engineering units.

As seen in FIG. 6 , the power supply board 100 converts 110VAC to 12 VDC and 5 VDC and is used to power the microprocessor 112 and lamp board 104 . The CCD board 106 generates a current proportional to the amount of light that strikes each pixel. The output of the CCD array is the input for the A/D board 108 . The A/D board 108 digitizes the signal and provides an output to the logic board 110 where filtering functions are performed. The program in the microprocessor determines which channel is on and switches the lamps on in sequence. The output of the system is a serial digital output 112 , i.e., RS-232 or 4-20 mA output

During calibration of the signal conditioner, the determination of the precise thickness of the wedge at each pixel along the CCD array is required. During the calibration of the sensor, calibration constants are entered into a file that are used to convert the gap into the appropriate engineering units, i.e., pressure or temperature. A calibration plot of sensor gap versus CCD pixel number is shown in FIG. 7 .

The method uses a two-dimensional (2-D) CCD array, e.g. 25 mm square, rather than the one-dimensional array used in the standard LASP described above. The 2-D array provides the capability to multiplex a large number of sensor signals from many fiber optic channels as shown in FIGS. 8 , 9 . Referring first to FIG. 8 , which shows the mapping of 64 optical fibers onto a 2-D CCD array. The cone shaped beam of light from each fiber is converted to a line shaped beam (beam line) of light by the cylindrical mirror shown in FIG. 9 . Each of the beam lines represents a separate channel and each beam line is projected onto the array at slightly different elevations corresponding to the difference in the vertical spacing between the fibers. The horizontal position of each beam line is staggered in the same manner as the fibers that deliver the light.

A cylindrical lens 42 could be substituted for the cylindrical mirror 36 defined above (see FIG. 10 ). The use of a lens instead of a mirror would double the length of the system shown in FIG. 9 and the lens would need to be corrected for chromatic aberration or other changes could be made to accommodate the wider beam line that would result from an uncorrected lens.

The polished fiber ends may be arranged in a variety of ways and held in a mounting block. In the preferred embodiment, sixty-four fibers are arranged side by side in two parallel V-blocks that are tilted at a slight angle as shown in FIG. 8 . The angle of tilt is determined to assure each beam line is offset from other beam lines such that, for example, the beam line from fiber 32 at the end of the first row does not interfere with the beam line from fiber 33 at the start of the next row.

FIG. 9 illustrates how light that exits from one fiber (point source) 32 is transformed from a cone 34 into a line 37 . The diverging light from the fiber fills the cylindrical mirror 36 and is reflected. In one dimension the light continues to diverge. In the other dimension, the light is focused into a narrow beam. The combination forms the beam line.

A Fizeau wedge 38 , e.g. 25 mm square, changes thickness in one dimension only and covers the CCD array 40 as observed in the top view of FIG. 9 . The ‘tear-drop’ shape 44 of the beam line 37 shown in the front view in FIG. 9 is a result of coma aberration. This aberration can be corrected by aspheric correction of the curvature of the mirror.

There is also a wavelength shift caused by chromatic dispersion through the wedge. It is necessary to correct for the chromatic dispersion in the signal processing algorithm.

The two-dimensional CCD array of the present invention permits multiplexing of a large number of sensor signals. This approach enables a significant increase in the update rate per channel compared to the standard LASP.