Linearization of mercury cadmium telluride photodetectors

Abstract

A method for linearization of photodetector response is provided that includes establishing one or more static calibration coefficients based on a comparison of a test photodetector response to a linear reference photodetector. In some examples, a dynamic calibration coefficient is determined based on an average photodetector signal. In some applications, such as FTIR, a single calibration coefficient is utilized to obtain a linearized ratio.

Description

Technical Field

[0001] The present disclosure relates to photodetector compensation. Background Art

[0002] Typical photodetectors for use at wavelengths in the short-wavelength infrared band (approximately 1.4 μm to 3 μm), the mid-wavelength infrared band (approximately 3 μm to 8 μm), the long-wavelength infrared band (8 μm to 15 μm), and the far-infrared band (15 μm to 1000 μm) exhibit a nonlinear response to input optical radiation. In some applications, such nonlinearities do not significantly limit the usefulness of the detector, but in other applications, even small deviations from a linear response can cause problems. For example, some applications require detection of optical radiation over a power range spanning six or more orders of magnitude. Any detector nonlinearity makes comparisons of small and large signal values ​​inaccurate. In applications such as Fourier transform spectroscopy, detector nonlinearities introduce or enhance frequency components in a manner that is independent of the properties of the sample being studied. Methods to address these and other limitations are needed. Summary of the Invention

[0003] The method includes directing a modulated light beam to a test photodetector and obtaining a modulation amplitude of an optical signal associated with detection of a first portion of the modulated light beam by the test photodetector. Based on the modulation amplitude of the optical signal from the test photodetector, at least one first calibration coefficient operable to linearize the test photodetector is determined and stored in a memory device. In some examples, a second portion of the modulated light beam is directed to a reference photodetector to obtain a modulation amplitude of an optical signal associated with detection of the second portion of the modulated light beam by the reference photodetector. Based on the modulation amplitudes of the optical signals from the test photodetector and the reference photodetector, at least one first calibration coefficient is determined.

[0004] A representative FTIR system includes a photodetector, a memory device storing at least one calibration coefficient associated with the photodetector, and a processor coupled to receive a light signal in response to illumination of the photodetector and linearize the light signal based on the at least one calibration coefficient. In some examples, the at least one calibration coefficient includes one or more of calibration coefficients a, b, and c, wherein the calibration coefficients are related to the measured light signal I. MEAS The associated linearized optical signal I LINEAR Produced as I LINEAR =aexp(bI MEAS In some examples, the processor is coupled to linearize the optical signal based on a reverse calculation of the received optical signal to a network node between the first amplifier and the second amplifier.

[0005] The foregoing and other features and advantages of the technology will become more apparent from the following detailed description, taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0006] Figure 1 A representative Fourier transform infrared spectrometer (FTIR) including linearized HgCdTe (MCT) photodetectors is shown.

[0007] Figure 2 A system for linearizing photodetectors based on reference photodetectors is shown.

[0008] Figures 2A to 2C Photodetector calibration data that can be acquired with a system such as that shown in Figure 2

[0009] Figure 3A A representative method of establishing static photodetector calibration coefficients and linearizing optical signals based on the calibration coefficients is shown.

[0010] Figure 3B A representative method of linearizing optical signals based on selected calibration coefficients is shown.

[0011] Figure 4 A representative method of establishing dynamic photodetector calibration coefficients is shown.

[0012] Figure 4A A representative beam modulation for dynamic calibration is shown.

[0013] Figure 4B and Figure 4C Data acquisition associated with dynamic linearization is shown.

[0014] Figure 4D A representative method of dynamic calibration is shown.

[0015] Figure 5 A representative spectral system including a processing system that provides linearized optical signals is shown.

[0016] Figure 6 A representative spectral system including a processing system that provides linearized optical signals and provides a current offset for reducing DC components of the optical signals prior to coupling for additional amplification is shown.

[0017] Figure 7 is a schematic diagram of a representative DC coupled amplifier suitable for providing optical signals for dynamic or static linearization.

[0018] Figure 8 is a schematic diagram of an alternative representative DC coupled amplifier suitable for providing optical signals for dynamic or static linearization.​

[0019] Figure 9 Representative methods for dynamic or static photodetector calibration based on back calculation to a reference node are shown.

[0020] Figure 10 A representative system for linearizing a photodetector using a variable optical attenuator or a variable light source is shown. DETAILED DESCRIPTION

[0021] As used herein, "sample photodetector" and "test photodetector" refer to the photodetector to be calibrated. In some cases, calibration of a single sample detector is sufficient for use with other photodetectors of the same configuration or type. Alternatively, each photodetector can be provided with its own unique calibration coefficients, which can be stored in a processor-readable memory device coupled to or provided with each photodetector. In some examples, a calibrated or inherently linear photodetector is used to compare and calibrate the response to beam modulation, and such a photodetector is referred to as a "reference photodetector." The reference photodetector can be a photodetector that is inherently suitably linear, or a previously calibrated test photodetector can be used as the reference photodetector. In the following examples, for ease of illustration, the calibration of an HgCdTe (MCT or mercury cadmium telluride) photodetector and the use of such a calibrated MCT photodetector are described. However, other ultraviolet, visible, or infrared photodetectors, or photodetectors for other ranges, can also be used. In some cases, a photodetector can operate linearly in some circuit connections and nonlinearly in other circuit connections, and calibration can be performed for any or all configurations as needed.

[0022] The disclosed methods are generally illustrated as using a reference photodetector to generate an optical signal that is used for linearization or other compensation of the photodetector under test. In other examples, a reference electrical signal can be provided to a light source or variable optical attenuator to generate a reference beam. In these examples, linearization or compensation can be based on the reference electrical signal. While a reference detector with a linear response facilitates calibration, any photodetector with known response characteristics can be used. In examples, the photodetector and associated electronics (e.g., an amplifier) ​​are said to generate a DC signal component or be DC-coupled. As used herein, such a photodetector and associated electronics generate an optical signal that includes a signal contribution associated with the average value of the received optical radiation. Therefore, the optical signal reflects both the average value and the spectrally induced modulation. In some examples, multiple amplifier stages are used, and in such examples, only the first stage (the stage coupled to the photodetector) must be DC-coupled so that a signal portion indicative of the DC component of the received optical radiation can be generated. If the DC component of the signal is completely removed, subsequent amplifiers can be AC-coupled. However, since the DC component is typically not perfectly removed, it may be preferable to DC-couple the entire amplifier chain. With appropriate measurement and processing as discussed below, subsequent amplifier stages do not need to be DC coupled. In the following examples, the average optical signal value associated with the optical signal measurement indicates the power delivered to the test photodetector and is referred to as E EFF In some examples, the test photodetector calibration coefficients are dynamic and are given as E EFF As used herein, a static calibration coefficient is a function of E EFF or other indication of the average optical signal without change; the dynamic calibration coefficient is the value obtained as E EFF or the value of another indicative function of the average optical signal.

[0023] One exemplary application of the disclosed method is FTIR. While MCT detectors are commonly used in FTIR, other detectors such as pyroelectric detectors based on deuterated triglycine sulfate, lithium tantalate, InGaAs, silicon or germanium bolometers, or other photodetectors can be similarly compensated if desired.

[0024] In the following, correction or compensation of the photodetector output signal is discussed with reference to linearization, i.e., compensating the photodetector output signal, or a signal based on the photodetector output signal, to be proportional to the detected optical power. Other types of compensation can be provided so that the photodetector output signal has other predetermined relationships with respect to the detected optical power. In the examples discussed below, linearization is discussed in which the compensated photodetector output signal or other signal is proportional to the detected optical power within a range of 5%, 2%, 1%, 0.5%, 0.1%, 0.01%, or less. Photodetector compensation is also referred to as calibration in the disclosed methods, and calibration or compensation parameters are obtained based on a specific model (exponential), but the calibration parameters can also be based on a second-order, third-order, or fourth-order polynomial, Bezier curve, or other function. In some of the examples below, calibration of a HgTeCd ("MCT") detector is described, but other detector types can be calibrated similarly. "Photodetector output signal" refers to a current, voltage, or combination thereof generated by a photodetector in response to received optical radiation. The term "signal" refers to a time-varying current or voltage generated in response to processing of the photodetector output signal. Such a signal may correspond to a photodetector output signal that has been amplified, filtered, or otherwise processed (as an analog signal or a digital signal). For convenience, the photodetector output signal and the signal responsive to the photodetector output signal (such as the processed photodetector output signal) are generally referred to herein as an "optical signal". In addition, although the signal refers to a time-varying current or voltage, such time-varying changes can be obtained and stored as a digital representation corresponding to a series of signal amplitudes stored in a processor-readable memory device. These digital signals are generally defined as representing the time variation of an analog signal. For example, a series of optical signal values ​​I(n) is also referred to as an optical signal with a time variation defined by n, where n=0, 1, ..., N-1, where N is a positive integer. As used herein, I may refer to optical power or an optical signal responsive to the detected optical power. In some examples, for convenience, the optical signal is represented as a voltage V because in many practical applications, the optical signal is processed as a voltage. However, depending on the type of photodetector and the circuit arrangement, the optical signal can be based on a light-induced current, voltage, or resistance.

[0025] Optical radiation refers to electromagnetic radiation propagating at wavelengths between approximately 100 nm and 1 mm. In some cases, optical radiation is said to propagate in a finite volume and is referred to as a beam. Optical power refers to the energy per unit time associated with a beam and is therefore proportional to the square of the associated electric field.

[0026] As indicated above, static calibration or compensation coefficients refer to values ​​that provide compensation (such as linearization) for the optical signal independent of the average value of the optical signal; dynamic calibration or compensation coefficients refer to values ​​that provide compensation (such as linearization) for the optical signal based on changes in the average optical signal (typically based on determining the dependence of one or more static calibration coefficients on the average optical signal).

[0027] Example 1 . Compensated detector FTIR

[0028] Reference Figure 1 , a representative FTIR system 100 includes a light source 102 that directs a light beam to a beam splitter 104, which delivers beam portions 105A and 105B to respective reflectors 106 and 108, which reflect beam portions 105A and 105B back to beam splitter 104 to form a combined light beam 105C. A sample 112 is positioned in the path of combined light beam 105C, which is incident on a detector 110, such as an MCT detector. A stage 114 is coupled to reflector 106 to vary the optical path length between beam portions 105A and 105B, typically by continuous or stepwise scanning. Scanning is initiated in response to processor-executable instructions stored in a portion 118 of a processor-readable storage device, which is provided as part of a control system 120, which also includes a CPU or other programmable or dedicated processor hardware (such as an FPGA, CPLD, or ASIC). In this example, control system 120 is coupled to photodetector 110 and receives light signal 111 associated with the scanning of reflector 106. In this example, optical signal 111 is digitized at analog-to-digital converter (ADC) 122 , and digital optical signal 123 is then linearized or compensated using processor-executable instructions and calibration parameters stored in portion 124 of the processor-readable storage device.

[0029] The digitized optical signal associated with a scan may be referred to as an interferogram, or a linearized or compensated interferogram if compensation has been applied. For a sample scan associated with N optical path length differences, the interferogram comprises a series of N optical signal values ​​IG for the N optical path length differences. S(n), n=0, ..., N-1. In some examples, digitization and linearization or other compensation processing can be performed before delivering the optical signal to the controller system 120. In some cases, the photodetector is part of a detection system that provides appropriate electrical bias (if necessary) to the photodetector 110 and includes an ADC to provide a digital optical signal to the control system 120. The linearization parameters can be stored by the controller 120 in a portion 124 of the processor-readable storage device and retrieved therefrom. The linearized optical signal 125 (interference pattern) is then Fourier transformed using processor-executable instructions stored in a portion 128 of the processor-readable storage device to produce a series of values ​​FFT K (IG S ), where K is an integer, usually K = 0, ..., N / 2-1, to correspond to I(G S In most applications, a background scan without a sample is performed and a corresponding background light signal (ie, background interferogram IG) is obtained. B (n)) and perform Fourier transform on it. Linearize the ratio of the corresponding components of the Fourier transform of the sample light signal and the background light signal FFT K (IG S ) / FFT K (IG B ) is obtained and provided as the FTIR system spectrum output.

[0030] The sample 112 is typically removable from the path of the combined light beam so that the sample 112 can be retrieved to perform a background scan. In some applications, an evacuated sample cell or a sample cell filled with a preferred material may be used. Although Figure 1 Transmission FTIR is shown, but in other examples, the beam reflects from the sample surface, is totally internally reflected in the sample, or interacts with the sample in some other way. Additionally, linearization of a detector (such as an MCT detector for FTIR) is only an example of the disclosed methods, and these methods can be used with other measurement systems, detectors, and in other wavelength ranges.

[0031] Example 2 . Linearization using static coefficients

[0032] In some examples, photodetector compensation (such as linearization) can be provided by selecting one or more static or dynamic calibration coefficients. For example, an exponential function can be used.

[0033]

[0034] To linearize the MCT photodetector,

[0035] where a cal 、bcal and c cal is the static calibration coefficient, f cal is the compensated (usually linearized) photodetector output signal, V mct is the uncompensated photodetector output signal. More generally, using such calibration coefficients, a range of optical signal values ​​I n (n=0, ..., N-1) can be linearized to produce a series of linearized optical signal values in

[0036]

[0037] Optical signal I n It may correspond to current, voltage, resistance, or other electrical characteristics in response to the light beam. n The photodetector signal may be a processed signal associated with amplification, filtering, or other processing applied to the photodetector signal. In some examples, linearization or other processing is applied to account for the effects of the amplification, filtering, or other processing. By performing calibration using the amplified photodetector signal, nonlinearities or other response characteristics associated with amplification or other photodetector processing can be compensated for along with the response characteristics of the photodetector.

[0038] Figure 2 A representative system 200 for linearizing an MCT photodetector 202 is shown. System 200 includes a controller 204 coupled to a variable light source 206 that generates a light beam 208 that is incident on a beam splitter 210. Beam splitter 210 directs respective portions of light beam 208 to the MCT detector 202 and a reference detector 212 that generates a reference light signal proportional to the optical power detected at the reference detector 212. Reference detector 212 can be a photodiode or other detector that generates a suitably proportional output, or a detector that generates a proportional output in combination with calibration circuitry. The optical signals generated by the MCT 202 and reference detector 212 are coupled to respective amplifiers 214, 216 or other amplification, buffering, or filtering circuitry, and the corresponding optical signals are delivered to controller 204.

[0039] The controller 204 is operable to vary the power of the light beam 208 and receive corresponding varying optical signals associated with the MCT detector 202 and the reference detector 212. In some examples, the power of the light beam generated by the variable light source 206 is periodically modulated (such as sinusoidally modulated) such that the light beam 208 is associated with a range of optical powers. The controller 204 may include an analog-to-digital converter and a processor-readable storage device such that the input optical signal is digitized and stored as a reference optical signal amplitude I REF and the MCT optical signal amplitude I MCT , such as Figure 2Aand Figure 2B As shown in . Figure 2A shows the essentially sinusoidal response of the reference detector 212 to the sinusoidal modulation of the light beam 208, while Figure 2B The corresponding response of the MCT detector to the same sinusoidal modulation is shown, which exhibits distortion. Figure 2C isI ref and I MCT Graphs plotted relative to each other illustrate the deviation of the optical signal generated by the MCT detector 208 from the linear response of the reference detector 212. For a linear response, I REF and I MCT The MCT detector can be linearized by changing the I MCT The curve is fitted to the linear response value I REF to provide, as discussed in detail below.

[0040] The beam power of the light beam 208 may be varied in steps, have other periodic or non-periodic modulations including sawtooth modulation and triangular modulation, and / or have a variable or fixed DC offset.

[0041] Example 3. Linearization using static calibration coefficients

[0042] Reference Figure 3A A representative method 300 for linearizing an optical signal to be proportional to detected optical power includes obtaining compensation data at 301, such as the I discussed above. REF , I MCT Although it is convenient for I REF is a linear function of the detector light signal, but other known responses may be used. At 302, a linearization function f is selected, such as an exponential, polynomial, Bezier curve, or function. A particular analysis format (such as a continuous function) is convenient so that compensation of an arbitrary light signal can be obtained for all values, but some or all of the selected data pairs (I ref ,I MCT ) to I MCT The selected value of and other values ​​determined, for example, by interpolation, provide linearization. At 304, any coefficients defining the compensation function are determined by curve fitting to f or otherwise processing the compensation data. For example, if the compensation function

[0043]

[0044] Selected (where I L (I MCT ) is related to I MCT associated linearized optical signal value), static calibration coefficient a cal 、b cal and c calDetermined by curve fitting, such as using a least squares fitting procedure to minimize Then, f(I MCT )=I L ((I MCT ) is the uncompensated optical signal I MCT At 306, the constant calibration coefficient a is set to cal 、b cal and c cal One or more of are stored in a memory device.

[0045] like Figure 3A As further shown in , a constant calibration factor a can be used cal 、b cal and c cal The representative method 307 of linearization includes receiving one or more optical signals of interest at 308 and linearizing the optical signals by using a calibration coefficient a at 310. cal 、b cal and c cal The optical signals are processed using the compensation function f, or a lookup table calculation corresponding to the calibration function and an interpolation process retrieved from a processor readable storage device. At 312, the linearized value I is output. L (I MCT ).

[0046] Example 4 Linearization using a reduced set of calibration coefficients

[0047] In some applications, the average optical signal value is not of interest, or the DC component has been removed by filtering. In these applications, the linearized optical signal I L Can be used as Produced because of c cal The relevant contributions are either removed by filtering or are not of interest. For these applications, only a cal and b cal , and only store and / or retrieve those values.

[0048] Further simplification is possible in systems using ratio-based measurements. In such systems, the optical signals associated with the sample under investigation and the reference are obtained, i.e., the optical signals associated with the sample are obtained. A series of optical signal values ​​and a series of optical signal values ​​associated with a reference The linearized sample and reference optical signal values ​​include the value a cal As a factor, so that in the ratio The value a is effectively removed from cal In any application of linearization, where and are the linearized optical signal values ​​associated with the sample and reference, respectively.

[0049] In some applications, the sample and reference optical signal values ​​are treated in a manner other than as simple ratios. For example, in an FTIR spectrometer, the optical signals associated with both the sample and the reference are Fourier transformed, and the sample spectrum of interest is obtained as the ratio of the corresponding components of the respective FFTs. In these applications, the calibration constant a is cal is also effectively removed by the FFT ratio and therefore does not need to be applied for linearization and does not need to be stored for such use.

[0050] For example, Figure 3B As shown in method 350 in FIG, at 352, a calibration process is selected based on one, two, or three calibration coefficients. At 354, the selected calibration coefficients are retrieved from a processor-readable storage device, and at 356, the selected calibration coefficients are used to linearize one or more input optical signals. At 358, the linearized signals are output.

[0051] Example 5 Dynamic calibration

[0052] The above constant (or static) calibration method can significantly reduce optical signal and / or photodetector nonlinearities, but does not compensate for heating or other changes in the optical signal or photodetector in response to the input light beam. For example, changes in the temperature of the photodetector can be associated with a shift in the calibration coefficient values. In some applications, such shifts are acceptably small and static calibration coefficient values ​​are sufficient. To account for shifts in the calibration coefficient values ​​in response to the input, the calibration coefficient values ​​can be refined based on the total optical signal. The total optical signal can be viewed as being proportional to the effective energy E applied to the photodetector. eff The total heat load is associated with the total heat load, which can be used as the N optical signal value I k The sum or average of

[0053]

[0054] The estimate E can be used eff Other methods are available, such as using optical signal values ​​obtained from previous measurements. This method is particularly suitable for FTIR using continuous scanning, where the signal value does not change rapidly between scans. Alternatively, in FTIR systems, the optical signal associated with optical delays far from zero path difference can be used as an approximation of the average value. However, in order to obtain E eff , a DC optical signal value must be obtained and the optical signal must include a DC component at at least one location in the processing, such that E effare available. As discussed above, the compensation can include effects due to the photodetection electronics (such as amplifiers and filters) in addition to just the photodetector. The photodetector signal values are obtained at suitable intervals to correspond to the current photodetector temperature or other photodetector or processing circuit effects associated with the measurement of interest. eff The photodetector signal values are obtained at suitable intervals to correspond to the current photodetector temperature or other photodetector or processing circuit effects associated with the measurement of interest.

[0055] As discussed above, when using an exponential function, the linearization can be based on one, two or three calibration coefficients. In one example discussed above, only the calibration coefficient b cal was used. This method is particularly suitable for applications where the DC or average value is not of interest and a ratio to a reference measurement value is used. For calibration using only the calibration coefficient b cal , the DC value is needed to determine E eff for dynamic compensation, but these DC values are otherwise not needed. In this example, the calibration coefficient b cal can be dynamically compensated for as:

[0056] b cal (E eff ) = a dbCal E eff + b dbCal ,

[0057] where a dbCal and b dbCal are dynamic calibration coefficients for dynamic compensation of the calibration coefficient b cal , and do not correspond to a cal and b cal used above. To determine b cal (E eff ), the values of these additional calibration coefficients must be estimated. In one method, a light beam with DC and modulated components can be directed to the detector being tested. Changing the DC component allows variation in E eff , and the modulated component allows determination of the calibration coefficients a cal , b cal and c cal at multiple values of E eff . Curve fitting such as discussed above can be used.

[0058] Referring to Figure 4 , a representative method 400 of determining one or more dynamic calibration coefficients includes selecting a light beam modulation amplitude and frequency at 402. These are typically selected to be at or near the range that the detector will be used. An average light beam power P AVE is selected at 404, and at 406 a light beam with the selected modulation and average light beam power is applied to a sample detector and a reference detector. Typical light beam modulations are in the range of 1 kHz to 1 MHz, and the average light beam power is in the range of 1 to 1000 microwatts. The light beam modulation is applied for a sufficient time to allow the detector to reach a steady state, and the light beam modulation is then removed. The reference detector is used to determine the DC value of the light beam, and the sample detector is used to determine the E Figure 4AAs shown in Figure 4A The average (or DC) value P at the modulation components 450 and 452 is shown. AVE At 408, multiple times t j Measure the optical signals IS(t j ), I REF (t j ). At 410, use I S (t j ) and I REF (t j ) together with the corresponding E EFF The average optical signal value of is used to determine one or more calibration coefficients as discussed above. At 414, the calibration coefficients and associated E are stored. EFF At 416 , it is determined whether calibration data at additional values ​​of beam power is needed. If so, the method returns to 404 to select a beam power. If not, dynamic calibration can be utilized at 418 to provide the selected calibration coefficient.

[0059] By changing the average optical power, the E EFF Values ​​get data sets, such as Figure 4B As shown, Figure 4B The E values ​​of the data sets 461-465 acquired with different average optical powers are shown. EFF Each data set provides the value of the associated E EFF A set of one or more calibration coefficients (i.e., a cal 、b cal and c cal one or more of ). Figure 4C E associated with data sets 461-465 is shown. EFF Various values ​​of E EFF The function of b cal and a linear fit 470 defining the dynamic calibration coefficients a stored at 420 dCal and b dCal .

[0060] In other examples, two calibration coefficients are changed dynamically. cal In addition, a cal Can be obtained dynamically as

[0061] a cal (E eff )=a daCal E eff +bd bCal ,

[0062] where a daCal and bdaCal is a dynamic calibration coefficient used to dynamically compensate for the calibration coefficients b S (t j ) and I REF (t j ) using the same calibration data I cal discussed above.

[0063] Figure 4D A representative method of dynamic linearization 480 is shown. At 482, one or more optical signals of interest are received, and at 484, an average optical signal value (corresponding to E EFF ) is obtained based on the received optical signals, optical signals from previous similar measurements, electrical offsets provided to one or more optical signal amplifier circuits based on average optical signal amplitudes, or otherwise. At 486, one, two, or more calibration coefficients can be retrieved from a lookup table in a processor-readable storage device, or calculated from the retrieved values by interpolation or otherwise. One or more of a cal , b cal , and c cal may be adjusted based on the average optical signal value. At 488, the received optical signals are linearized using at least one dynamic calibration coefficient associated with the average value of the received optical signals.

[0064] Example 6. Representative FTIR measurements

[0065] In FTIR, optical signals associated with a sample and optical signals associated with a background are obtained such that the measured spectrum reflects the sample and not any extraneous effects included in the background. In FTIR, linearization can be performed without a full set of calibration coefficients. For example, a background interferogram vector b = b0,..., b N-1 and a sample interferogram vector s = s0,..., s N-1 , where each is a series of N optical signal values. As discussed above, calibration coefficients can be used to correct the background and sample interferograms. The symbol “^” is used to denote corrected (i.e., linearized) values, such that the nth element of the linearized background and sample interferogram vectors are:

[0066]

[0067] and

[0068]

[0069] The average can be removed by subtraction; the resulting vector and elements are denoted with the symbol “~”, such that

[0070] and

[0071]

[0072] These simplify to:

[0073] in

[0074]

[0075] as well as

[0076] in

[0077]

[0078] Using capital letters to denote the Fourier transform, i.e., as vectors B and S, the kth component of the Fourier transform of the linearized zero-mean background interferogram and the linearized zero-mean sample interferogram is:

[0079]

[0080] and

[0081]

[0082] In FTIR, the ratio T k =S k / B k is the quantity of interest, and obviously the common factor is a cal Therefore, in FTIR, the linearization of the background interferogram to remove the mean value and the sample interferogram can be performed as: where i k is the kth component of the interference pattern or its Fourier transform.

[0083] Example 7. Representative Spectrometer

[0084] Reference Figure 5 , the compensated spectrometer system 500 includes a spectrometer 502 that couples a portion of a light beam associated with a spectral component to a detector 504 (such as an MCT detector). The detector provides an optical signal to a DC-coupled amplifier 506, which provides an amplified and / or filtered optical signal to a processing system 508, which may include an ADC (not shown), a processing device, and a processor-readable storage device including a processor for linearizing, determining E EFF , storing constant and / or dynamic calibration portions 510, 512, 514 to provide a linearized output using dynamic or constant calibration using one, two or three calibration coefficients.

[0085] Figure 6 A compensated spectrometer system 600 is shown, which includes a Figure 5 The spectrometer is similar to the spectrometer 602 and a first amplifier (DC coupled amplifier) ​​606 coupled to a photodetector 604 (such as an MCT photodetector). The amplifier 606 is coupled to a second amplifier 608 to direct the optical signal to a processor system 610. In this example, the processor system 610 is coupled to establish a fixed output of the photodetector 604 when the photodetector is not exposed to any light beam, thereby reducing the apparent signal contribution due to dark current. If necessary, the processor system 610 also provides one or more control signals to the amplifier 606 via a digital to analog converter (DAC) 614 to select the amplifier gain and offset using processor executable instructions stored in a memory device. By adjusting the offset and gain, the amplifier range can be used for the AC portion of the optical signal and the associated digital optical signal can use the full range. The optical signal from the DC coupled amplifier 606 and any offset provided by the ADC 614 can be used to establish E for dynamic calibration. EFF The processing system 610 may store calibration coefficients and use them for linearization, determination of E EFF , dynamic calibration, and processor-executable instructions for selecting one, two, or three calibration coefficients.

[0086] Example 8 . Representative photoelectric signal amplifier

[0087] Reference Figure 7 , a representative amplifier 700 for use with a photodetector 702 includes a first operational amplifier 706 and a resistor 708 arranged as a transimpedance amplifier based on the photocurrent received from the photodetector 702. The resistance of the resistor 708 can be selected to provide a suitable voltage output for any detected light signal. The first operational amplifier 706 provides the light signal to a second operational amplifier 710, which provides a light signal voltage output V OUT The first operational amplifier 706 is also coupled to a peak detector 712 that provides an indication of the peak optical signal amplitude from the first operational amplifier 706 as a digital signal to a processor 714. The processor 714 is coupled to a DAC 716 and a digital potentiometer 718 that is coupled to the digital controller resistor R POT The offset and gain provided by operational amplifier 710 are established; the DAC output is coupled to buffer amplifier 720 and to digital potentiometer 718 via resistor 722; the gain is based on the ratio R POT / R2 (for purposes of explanation, the dashed line indicates the digital signal path.) The offset can be set to have a small amplitude, and the gain set so that the optical signal amplitude matches the available digital and / or analog voltage range, including the range associated with any downstream ADC, although additional gain stages can be used if desired.

[0088] The processor 614 is also coupled to provide a digital offset signal to a DAC 726, which provides an analog offset signal to a buffer amplifier 728. This digital offset signal is correlated with the photodetector signal in the absence of an input light signal and is used to at least partially remove dark current from the light signal received at the first operational amplifier 706. The photodetector 702 is coupled to a supply voltage V s , and the resulting associated bias current 730 can be offset with a suitable digitally controlled offset voltage.

[0089] As disclosed above, the output signal V OUT Typically digitized and linearized. In some examples, the transimpedance amplifier formed by operational amplifier 706 and resistor 708 is typically sufficiently linear that any nonlinearity at reference node 732 is due to photodetector 702. The gain between the photodetector and the reference node is fixed; the calibration coefficient can be measured relative to reference node 732. Output V OUT is a function of the digital gain and offset signals applied to the digital potentiometer 718 and DAC 716, respectively, so that the output signal V OUT can be calculated in reverse to provide the optical signal value at the reference node 732 for linearization. The linearization can be based on the node with which the calibration data is associated.

[0090] like Figure 7 As shown, the output voltage V OUT By V OUT =V OS G OS +V RN G SIG Given, where V OS is the offset voltage (i.e., the voltage provided by DAC716), G OS is the offset gain (i.e., the gain from the output of DAC 716 to the output of op amp 710), V RN is the voltage at the reference node 732, and G SIG is the signal gain from reference node 732 to the output of operational amplifier 710. It can be seen that G OS and G SIG By and Given. Resistor R POT is the digital control signal NPOT function, so that G OS and G SIG The reference node voltage V RN Then:

[0091]

[0092] The linearization discussed above is applied to V using one or more static and / or dynamic calibration coefficients obtained for reference node 732. RN .

[0093] DAC 716 receives a control signal (such as a control word N DAC ), and the digital potentiometer 718 receives a control signal (such as a control word N) from the processor 714. POT ) to establish V OS and R POT In one example, the control word will be V OS and R POT Established as and where R MAX is the maximum resistance available, m is the bit resolution of the digital potentiometer 718, V REF is the reference voltage of the DAC, and k is the bit resolution of the DAC 716. Using these digital signals, reverse calibration can be performed. In addition, V OS and R POT , to reduce V OUT The DC component in φ and sets the gain provided by the second operational amplifier 710.

[0094] Linearization can be performed by processor 714 or other processors and if gain and offset are to be calculated elsewhere, V is provided along with other parameters as needed. OUT 、N POT and N DAC .

[0095] Example 9 Representative photoelectric signal amplifier with current injection

[0096] In many examples, a DC-coupled amplifier is required for linearity correction because the DC component of the modulated optical signal incident on the photodetector is large. However, in FTIR measurements, spectroscopists have little analytical interest in the DC component. Therefore, in an alternative amplifier configuration, in addition to the offset V discussed above, OSIn addition, a DC component with an opposite amplitude to the DC optical signal component is introduced into the signal chain. This introduced DC component can ensure that the amplifier stage (such as the first stage) is not saturated. The first stage amplifier is usually a transimpedance amplifier because of its excellent linearity, but such amplifiers usually have limited current drive capability, which limits the transimpedance amplifier feedback resistor (e.g., Figure 7 How small R1 in can be. This in turn means that there is a minimum practical gain for the first stage. Combined with a large DC optical signal component with a small superimposed interferogram, amplifier saturation is possible, especially for hot samples where an additional DC component is generated.

[0097] Figure 8 The amplifier system 800 shown in FIG can solve the problem of possible transimpedance amplifier saturation. The amplifier system 800 is similar to Figure 7 The amplifier system includes a current source 802, which is controlled by a DAC 816 to convert the current I SET Together with the photodetector current, it is injected into the operational amplifier 706. If it is not necessary to introduce this current offset into the photodetector current, the current injection can be disabled. SET cancels the DC part of the photocurrent and is given by the product R1I before applying linearization SET is added to the voltage V RN (i.e., V referenced back to reference node 732 OUT ). In this configuration, V OS Typically it can be set to 0. The amplifier system 800 is shown as providing a linearized output from a processor, but linearization can be provided at a remote processing system that receives V OUT and N DAC 、N POT 、V REF and I SET Or other parameters required for reverse calculation to reference node 732.

[0098] Example 10 . Representative linearization methods with back calculation

[0099] Reference Figure 9 Method 900 includes measuring an output optical signal at 902 and back-calculating the optical signal amplitude to a reference node at 904. The back-calculation can be based on a gain and / or offset applied to the photodetector signal. At 906, a linearization method and corresponding linearization coefficients are selected, such as those associated with single-parameter or multi-parameter static or dynamic compensation. At 908, calibration coefficients are retrieved, and at 910, the back-calculated optical signal is linearized, and at 912, the linearized back-calculated optical signal is stored. In an FTIR measurement, an FFT of the back-calculated optical signal is calculated and stored or displayed.

[0100] Example 11 Calibration using a variable light source or optical attenuator

[0101] Reference Figure 10 The representative photodetector calibration system 1000 includes a variable light source 1006 and / or a variable optical attenuator 1008 coupled to a processor system 1010, the processor system being operable to respectively utilize corresponding control signals (such as a control voltage V LS 、V ATTN ) to change the optical power in the light beam 1010. The light beam 1010 is directed to a test photodetector 1012, which generates an optical signal that is directed to an amplifier 1014, and the corresponding amplified optical signal is coupled to a processor 1014, typically an ADC, to provide a digital optical signal having a power that can be compared to V LS or V ATTN The values ​​stored in the memory device processor 1024 together with the correlation values ​​can be used in any calibration process disclosed above.

[0102] Representative Examples

[0103] Embodiment 1 is a method comprising: directing a modulated light beam to a test photodetector; obtaining a modulation amplitude of an optical signal associated with detection of a first portion of the modulated light beam by the test photodetector; determining at least one first calibration coefficient operable to linearize the test photodetector based on the modulation amplitude of the optical signal from the test photodetector; and storing the at least one first calibration coefficient in a memory device.

[0104] Embodiment 2 includes the subject matter of embodiment 1, and further includes: directing a second portion of the modulated light beam to a reference photodetector; obtaining a modulation amplitude of an optical signal associated with detection of the second portion of the modulated light beam by the reference photodetector; and determining the at least one first calibration coefficient based on the modulation amplitude of the optical signal from the test photodetector and the reference photodetector.

[0105] Example 3 includes the subject matter of any of Examples 1-2, and further specifies that the modulation of the light beam is established based on variable activation of a light source that produces the light beam or a variable attenuator placed in the path of the light beam from the light source, and that the at least one calibration coefficient is determined based on the modulation amplitude of the light signal from the test photodetector and the variable activation or variable attenuation of the light source.

[0106] Embodiment 4 includes the subject matter of any of embodiments 1-3, and further specifies that the optical signal is a photodetector signal.

[0107] Embodiment 5 includes the subject matter of any of embodiments 1-4, and further specifies that the at least one first calibration coefficient includes only one first calibration coefficient.

[0108] Embodiment 6 includes the subject matter of any of embodiments 1-5, and further specifies that the at least one first calibration coefficient comprises three first calibration coefficients associated with a linearization based on an exponential function with an offset.

[0109] Embodiment 7 includes the subject matter of any of embodiments 1-6, and further specifies that the at least one first calibration coefficient comprises calibration coefficients a, b, c, wherein the linearized optical signal is generated as I LINEAR =aexp(bI MEAS )+c, where I LINEAR is the measured optical signal I MEAS associated with the linearized optical signal.

[0110] Embodiment 8 includes the subject matter of any one of embodiments 1-7, and further specifies that the at least one first calibration coefficient comprises a first calibration coefficient b, wherein the linearized optical signal is generated as I LINEAR =exp(bI MEAS ), where I LINEAR is the measured optical signal I MEAS associated with the linearized optical signal.

[0111] Embodiment 9 includes the subject matter of any of embodiments 1-8, and further specifies that the first calibration coefficients a, b, c are constants.

[0112] Example 10 includes the subject matter of any one of Examples 1-9, and further includes: changing the average power of the modulated light beam directed to the test photodetector; determining at least one second calibration coefficient based on the changing average power of the modulated light beam, wherein the second calibration coefficient is at least one dynamic calibration coefficient associated with the average power dependence of a selected one of the first calibration coefficients; and storing the at least one second calibration coefficient in the memory device.

[0113] Embodiment 11 includes the subject matter of any of embodiments 1-10, and further specifies that the at least one second calibration coefficient is associated with the calibration coefficient b.

[0114] Embodiment 12 includes the subject matter of any one of embodiments 1-11, and further specifies that the at least one dynamic calibration coefficient includes two calibration coefficients A and B, such that the calibration coefficient b=A(E eff )+B, where E eff is associated with the average power of the modulated light beam.

[0115] Embodiment 13 includes the subject matter of any one of embodiments 1-12, and further includes linearizing the measured optical signal to I LINEAR =exp(bI MEAS ), where I LINEAR is the measured optical signal I MEAS associated with the linearized optical signal.

[0116] Embodiment 14 is an FTIR system comprising: a photodetector; a memory device storing at least one calibration coefficient associated with the photodetector; and a processor coupled to receive an optical signal in response to illumination of the photodetector and linearize the optical signal based on the at least one calibration coefficient.

[0117] Embodiment 15 includes the subject matter of embodiment 14, and further specifies that the at least one calibration coefficient comprises one or more of calibration coefficients a, b, c, wherein the calibration coefficients a, b, c are related to the measured optical signal I MEAS The associated linearized optical signal I LINEAR Produced as I LINEAR =aexp(bI MEAS )+c.

[0118] Embodiment 16 includes the subject matter of any of Embodiments 14-15, and further specifies that the at least one calibration coefficient comprises a calibration coefficient b, wherein MEAS The associated linearized optical signal I LINEAR Produced as I LINEAR =exp(bI MEAS ).

[0119] Embodiment 17 includes the subject matter of any of Embodiments 14-16, and further specifies that the memory device stores at least one dynamic calibration coefficient, wherein the processor linearizes the optical signal based on the at least one calibration coefficient and the at least one dynamic calibration coefficient.

[0120] Embodiment 18 includes the subject matter of any of embodiments 14-17, and further specifies that the at least one dynamic calibration coefficient is associated with the calibration coefficient b.

[0121] Embodiment 19 includes the subject matter of any one of embodiments 14-18, and further specifies that the at least one dynamic calibration coefficient includes two dynamic calibration coefficients A and B, such that the calibration coefficient b=A(E eff )+B, where E eff is associated with the average power of the modulated light beam.

[0122] Example 20 includes the subject matter of any one of Examples 14-19, and further includes linearizing the measured optical signal to I LINEAR =exp(bI MEAS ), where I LiNEAR is the measured optical signal I MEAS associated with the linearized optical signal.

[0123] Embodiment 21 includes the subject matter of any of Embodiments 14-20, and further includes a DC-coupled amplifier coupled to the photodetector.

[0124] Embodiment 22 includes the subject matter of any one of Embodiments 14-21, and further specifies: the amplifier includes a first amplifier and a second amplifier, wherein the first amplifier is a DC amplifier coupled to the photodetector; and the processor is coupled to provide a variable gain and offset to the second amplifier, and linearizes the optical signal based on the at least one calibration coefficient and the variable gain and offset.

[0125] Embodiment 23 includes the subject matter of any one of Embodiments 14-21, and further specifies that Eeff is determined based on the measured optical signal or the offset applied to the second amplifier.

[0126] Embodiment 24 includes the subject matter of any of Embodiments 14-23, and further provides that the processor is coupled to linearize the optical signal based on a reverse calculation of the received optical signal to a network node between the first amplifier and the second amplifier.

[0127] Embodiment 25 includes the subject matter of any of Embodiments 14-25, and further includes a digital potentiometer and a digital-to-analog converter coupled to the processor and the second amplifier to establish the variable gain and offset.

[0128] Embodiment 26 includes the subject matter of any of Embodiments 14-25, and further comprising a current source that provides current to the photodetector based on an average photocurrent generated in response to the illumination of the photodetector.

[0129] Embodiment 27 includes the subject matter of any one of embodiments 14-26, and further specifies that the at least one dynamic calibration coefficient includes two dynamic calibration coefficients A and B, such that the calibration coefficient b=A(E eff )+B, where E eff is associated with the average power of the modulated light beam.

[0130] Embodiment 28 includes the subject matter of any of Embodiments 14-27, and further comprising a current source that provides current to the photodetector based on a photodetector dark current.

[0131] In view of the many possible embodiments to which the principles of the disclosed invention may be applied, it should be recognized that the illustrated embodiments are only preferred examples and should not be taken as limiting the scope of the disclosure.

Claims

1. A method comprising: directing the modulated light beam to a test photodetector; obtaining a modulation amplitude of an optical signal associated with detection of a first portion of the modulated optical beam by the test photodetector; determining at least one first calibration coefficient operable to linearize the test photodetector based on the modulation amplitude of the optical signal from the test photodetector; as well as The at least one first calibration coefficient is stored in a memory device.

2. The method according to claim 1, further comprising: directing a second portion of the modulated light beam to a reference photodetector; obtaining a modulation amplitude of an optical signal associated with detection of the second portion of the modulated optical beam by the reference photodetector; as well as The at least one first calibration coefficient is determined based on the modulation amplitudes of the optical signals from the test photodetector and the reference photodetector.

3. A method according to claim 1, wherein the modulation of the light beam is established based on variable activation of a light source generating the light beam or a variable attenuator placed in the path of the light beam from the light source, and at least one calibration coefficient is determined based on the modulation amplitude of the light signal from the test photodetector and the variable activation or variable attenuation of the light source. The method of claim 1 , wherein the optical signal is a photodetector signal. The method of claim 1 , wherein the at least one first calibration coefficient comprises only one first calibration coefficient. The method of claim 1 , wherein the at least one first calibration coefficient comprises three first calibration coefficients associated with a linearization based on an exponential function with an offset.

7. The method of claim 1 , wherein the at least one first calibration coefficient comprises calibration coefficients a, b, c, wherein the linearized optical signal is generated as I LINEAR =aexp(bI MEAS )+c, where I LINEAR is the measured optical signal I MEAS associated with the linearized optical signal.

8. The method of claim 1 , wherein the at least one first calibration coefficient comprises a first calibration coefficient b, wherein the linearized optical signal is generated as I LINEAR =exp(bI MEAS ), where I LINEAR is the measured optical signal I MEAS associated with the linearized optical signal.

9. The method of claim 7, wherein the first calibration coefficients a, b, c are constants.

10. The method according to claim 1, further comprising: varying an average power of the modulated light beam directed to the test photodetector; determining at least one second calibration coefficient based on a varying average power of the modulated light beam, wherein the second calibration coefficient is at least one dynamic calibration coefficient associated with an average power dependence of a selected one of the first calibration coefficients; as well as The at least one second calibration coefficient is stored in the memory device. The method of claim 10 , wherein the at least one second calibration coefficient is associated with the calibration coefficient b.

12. The method of claim 11, wherein the at least one dynamic calibration coefficient comprises two calibration coefficients A and B, such that the calibration coefficient b=A(E eff )+B, where E eff is associated with the average power of the modulated light beam.

13. The method according to claim 12, further comprising linearizing the measured optical signal to I LINEAR =exp(bI MEAS ), where I LINEAR is the measured optical signal I MEAS The linearized optical signal is associated.

Images