Electromagnetic detection apparatus

ES 2 329 754 T3

DESCRIPTION

Electromagnetic detection device.

Technical scope

The present invention relates to electromagnetic detection apparatus and, in particular, to an electromagnetic probe for measuring a parameter of a sample, especially an infrared absorption probe.

Previous technique

Infrared absorption probes are well known and are used for example to measure constituents of samples (for example the moisture content of paper or tobacco, or the fat, protein or water content of food), the amounts of substances absorbed on a substrate, the thickness of coatings or films on a substrate, or the degree of aging of resins on a printed circuit board. In this specification, the term "parameter" is used to denote the property (composition, coating thickness, etc.) of the sample being measured.

Infrared absorption probes conventionally operate by projecting infrared radiation at two or more wavelengths onto a sample or substrate and measuring the intensity of the radiation reflected, transmitted, or scattered by the sample. Signals proportional to the measured intensity are processed to provide a value of the parameter being measured. At least one or the two or more wavelengths projected by the probe is chosen to be absorbed by the parameter of interest while the -1 of the other wavelength is chosen to be absorbed by the parameter of interest while at least one of the other wavelengths is chosen so as not to be substantially affected by the parameter of interest. For example, when measuring the amount of water in a sample, one of the wavelengths (the "measurement wavelength") can be chosen at a water absorption wavelength (either 1.45 microns or 1 , 94 microns) and the other wavelength (known as the "reference wavelength") is chosen to be one that is not significantly absorbed by water.

Generally, the probes include a source of infrared radiation having a broad emission spectrum and a detector to receive the radiation reflected, scattered or transmitted by the sample; Filters are placed between the source and the sample to expose the sample only to the desired reference and measurement wavelengths; in this case, the sample is successively exposed to radiation at the selective wavelengths, for example by placing appropriate filters on a rotating wheel in front of the radiation source. Alternatively, the filter wheel can be positioned between the sample and the detector and each filter is interposed successively between the sample and the detector. Naturally, if the source can produce radiation of the desired wavelength without the use of filters, then such filters can be dispensed with.

The detector measures the intensity of the light after the interaction with the sample and produces a signal according to the intensity of the incident radiation behind it. In the simplest case, by calculating the relationship between the signal from the detector when it receives light at the measurement wavelength versus when it receives light at the reference wavelength, a signal can be obtained that provides a measurement of the affected parameter. , for example the amount of moisture in the sample. Often, several measurement wavelengths and / or several reference wavelengths are used and the signals from the measurement wavelengths and the reference wavelengths are used to calculate the affected parameter.

The detectors normally used in such measurement probes are conventionally lead sulfide (PbS) detectors, as they show better detection ability and wavelength response than most other detectors that can be used in such applications. . However, PbS detectors have several limitations, particularly including the following:

(a) Temperature sensitivity: the resistance of a typical detector drops 25% for every 10 ° C increase in temperature.

(b) Non-linearity: The detector's response to incident radiation is non-linear throughout the detector's operating range.

(c) Response time: the response time of the detector usually limits the rate at which different wavelengths can be detected, that is, the rate at which successive filters can be used. Faster filter data rates tend to cause the signal of the wavelength obtained from one filter to lag so far that it fades into it from the wavelength obtained from the next filter, thereby causing "interference".

(d) Noise: at low frequencies of detector operation, a type of noise known as 1 / f noise predominates. If a relatively low filter data rate is chosen to avoid interference, then such noise becomes a problem.

ES 2 329 754 T3

It is clear from the foregoing that the detectors currently used in measurement probes suffer from several drawbacks, not the least of which is their response time.

In a different field, GB-1401699 relates to an electromagnetic measuring apparatus for measuring the gap between the rollers of a steel rolling mill. In this prior apparatus, a main source of illumination is caused to emit a fluctuating beam of light through the use of a light modulator. Modulated light passes through an aperture and between two workpieces and received through an aperture in a detector unit. The unit includes a photoelectric transducer to convert the received light beam into an electrical signal having the frequency as the modulation generated by the light modulator. The alternating electrical signal is applied to a switching stage to control an adjusting device to alter the distance between the work pieces.

The present invention seeks to address these problems and improve the performance of detectors used in infrared measurement probes.

Presentation of the invention

In accordance with the present invention, there is provided an infrared probe for measuring a parameter of a sample, as defined in claim 1.

In a preferred form of the invention, the means for stabilizing a characteristic response of the detector is arranged to achieve at least one of the following: linearization of the response of the detector and / or extension of the frequency of the response of the detector.

In an infrared measurement probe, improving linearity tends to improve stability with the temperature of the measurement probe and leads to a more predictable calibration. Extending the frequency response of the detector allows the use of faster rate of filter data, and thus reduces the response time of the measurement probe. Faster filter speeds also result in reduced ambient light sensitivity and can lead to improvements in noise.

Advantageously, the means for controlling the radiation source are arranged to adjust the intensity of the radiation emitted by this source.

In one embodiment of the invention, the control means is arranged to adjust the intensity of the radiation emitted by the radiation source to keep the detector signal at a substantially predetermined level.

In another embodiment of the invention, the control means is arranged to adjust the intensity of the radiation emitted by the radiation source to minimize variations in the detector signal.

Advantageously, a feedback path is provided from the detector circuit output to the second radiation source for this purpose.

Short description

The invention is further described, by way of example, with reference to the accompanying drawings, in which:

Figure 1 is a schematic section through the head of a known probe, for purposes of explanation;

Figure 2 is a detailed schematic section through a detector mounting plate of said infrared probe, but illustrating the present invention;

Figure 3 is a block diagram showing an example of a feedback loop not within the scope of the claims;

Figure 4 is a block diagram showing another embodiment of the present invention;

Figure 5 is a graph showing the non-linearity of the response of a conventional PbS detector.

Figure 6 is a graph showing the frequency response of a PbS detector in a known probe; ^Y

Figure 7 is a graph showing the frequency response of a detector when the present invention is employed.

Methods of carrying out the invention

Referring initially to Figure 1, this shows the head 10 of a known infrared probe, for example as described in our published PCT application WO98 / 22806. Head 10 contains a lamp 12

ES 2 329 754 T3 that provides a source of infrared radiation, and a circular filter wheel 14 driven by a motor 16. The filter wheel 14 carries a series of filters, for example 5 filters, and each filter is designed to pass a different selected emission wavelength. The light passed through the respective filters is directed towards a detector mounting table 18, as described below.

The mounting table 18 carries a beam splitter 20 that reflects a part of the light beam downward out of the infrared probe 10 towards a sample 22. A remaining part of the infrared light beam striking the beam splitter 20 is reflected within the beam splitter to a detector assembly 24 that includes a photoelectric sensor. Meanwhile, the light emitted by the head 10 towards the sample 22 is reflected back from the sample 22 towards the collecting mirror 26 in the head 10 and thence to another detector assembly 28 that includes another photoelectric sensor. The two detector assemblies 24, 28 thus generate detection signals representing, respectively, the intensity of the light emitted by the lamp 12 and filtered by the selected filter, and the intensity of that same light after reflection from the sample 22 The detector assembly 28 is normally referenced as a secondary detector assembly. The signals generated by the two detector assemblies 28, 24 are processed in a known manner to provide a measurement of a parameter of the sample 22.

The measurement probe written so far is known as a backscatter probe because the light that is detected is scattered back from the sample whose parameter is to be measured. Another known measuring probe detects the light that is transmitted through a sample whose parameter is to be measured.

Turning now to Figure 2, this shows a detector arrangement in accordance with the present invention having a mounting board 18 carrying a primary detector assembly 28 and a secondary detector assembly 24 as shown in Figure 1. As shown You know, the primary detector assembly 28 incorporates a PbS detector 282 and a blocking filter 284, which is a small sheet of silicone, mounted on the front of the detector 282 to block visible light. Similarly, secondary detector assembly 24 incorporates a PbS detector 242 and a blocking filter 244, also a small silicone sheet, mounted on the front of the detector to block visible light.

In accordance with the invention, however, mounting board 18 also supports additional sources of radiation, in this example a first Gallium Arsenide (GaAs) infrared emitter or diode 30 directed toward blocking filter 244 and a second diode of GaAs 32 directed toward blocking filter 284. GaAs diode 30 directs infrared radiation to blocking filter 244, and such radiation is reflected back by blocking filter 244 to mounting board 18 and radiates to detector 242 in use. Similarly, GaAs diode 32 directs radiation toward blocking filter 284 and that radiation is reflected back toward mounting board 18 and radiates e detector 282 in use. Consequently, during operation of the measurement head 10, the primary and secondary detectors 282, 242 are receiving not only the successively applied measurement and reference wavelengths, but also radiation from the two GaAs diodes 32, 30. The two GaAs diodes 32,30 are arranged to cooperate with the detectors 282, 242 in such a way as to stabilize the outputs of these detectors, and for this purpose the intensity of the radiation emitted by the gods 32, 30 is controlled in a as described later.

Figures 3 and 4 show different embodiments of arrangements for controlling diodes 32, 30 and thus for stabilizing the outputs of detectors 282, 242. For the sake of simplicity, only the arrangements for diode 32 will be described in each case. and detector 282, but it is to be understood that the same arrangement will be employed for diode 30 of detector 242.

Referring first to Figure 3 which shows an example, which shows an example not within the scope of the claims, the detector 282 is arranged to generate an output signal, which is amplified by an inverting amplifier 34 and supplied as a voltage output signal to an output 36. A proportion of the signal supplied to output 36 is sent backward via feedback circuit 37 to diode 32. More particularly, the feedback circuit 37 includes a summing amplifier 38 having one input connected to receive the voltage signal sent backward from the output 36 and another input arranged to receive a predefined voltage supplied on a line 39. The output of the summing amplifier 38 is connected to a voltage-to-current converter 40 whose output controls diode 32.

When the magnitude of the output signal from the detector 282 begins to increase, due to an increase in the intensity of light at the reference or measurement wavelength received there, the current supplied to the GaAs diode 32 begins to drop to reduce the intensity of the illumination emitted by diode 32. The feedback circuit 37 thus acts as a negative feedback loop and thus seeks to maintain the total elimination that radiates to the detector 282, that is, the combined illumination that is reflected back from the sample 22 and that is emitted by the GaAs diode. 32, at a substantially constant predetermined level.

The predetermined level is set to correspond to the maximum level of external illumination expected to be encountered by the measuring probe in service, and is obtained by calibrating the GaAs diode 32, by appropriate selection of the bias voltage of line 39, to emit a flux corresponding to this radiation under conditions when there is no radiation with a reference or measurement wavelength will fall on detector 282. The effect of this is to confine the detector 282 to a particular part of its characteristic operating response and thus greatly reduce changes in the total radiation incident from the detector 282.

ES 2 329 754 T3

The voltage output signal that is obtained at the output 36 effectively corresponds to an error signal that represents the difference between the predetermined level and the current level of radiation currently incident on the detector 282. This difference in turn represents the amount of light that can be reflected back from sample 22. Said error signal is then processed in a known manner to provide an indication of the parameter to be measured.

Figure 4 shows an embodiment of the present invention that is an improvement on the arrangement shown in Figure 3, in which the detector 282 is not required to operate in said saturated state. Similar parts are designated by the same reference numerals as Figure 3.

In the circuit illustrated in Figure 4, the detector 282 is connected to the amplifier 34 by means of a high pass filter 42. The filter 42 is arranged to pass signals at the frequencies normally generated in use of the measurement probe by means of selection of a desired filter data rate. At the same time, however, filter 42 is arranged to filter out any low frequency variations generated in response to very gradual changes, for example, in ambient temperature or steady state ambient lighting.

In this embodiment, the bias voltage applied to line 39 is set to maintain the level of radiation falling on detector 282 at an amount that represents the largest expected flux variation in service. This level will naturally be considerably lower than the saturation level used in the embodiment of figure 3. The present embodiment thus seeks to simply minimize variations in the total elimination radiated by the detector 282, rather than maintaining the total illumination level at a fixed predetermined level.

As before, the voltage signal obtained at the output 36 constitutes an error signal that represents the difference between the preset level and the current level of incident radiation from the detector 282. The difference represents the amount of light that falls on the detector 282 that is reflected back from sample 22.

The circuit illustrated in Figure 3 results in the detector 282 operating in a predetermined region of its characteristic operating response, while the circuit illustrated in Figure 4 causes the detector 282 to operate in a restricted range throughout its performance characteristics. In both cases, experiments have shown that the detector benefits from improved linearity and an extended frequency response.

Figure 5 shows a graph representing the response of a conventional detector dependent on incident radiation for the full range of detector operation. The optical power of the incident radiation is plotted along the X axis and the normalized response of the detector is plotted along the Y axis. As shown, the response of the detector is 0.072 to 5 pW of incident radiation and is 0.141 to 10 pW of incident radiation, and therefore in this region of the curve the response of the detector is almost linear. However, making the same comparison for 50 pW and 100 pW of incident radiation, the detector response is in the 0.589: 1,000 ratio, which is far from linear.

The curve shown in Figure 5 shows that the smaller the change in radiation incident on the detector, the more linear the detector signal.

Consequently, as the feedback arrangement according to the present invention reduces changes in the overall incident ratio at the detector, the resulting signals generated by the detector will be more linear.

This is especially advantageous in situations where the detector is brightly illuminated during operation and deep absorptions are likely, as then a linear response is essential to ensure proper calibration and good temperature stability.

Figure 6 shows a frequency response graph for a conventional PbS detector on a known measurement probe at room temperature. The normalized response of the detector is plotted against the cutoff frequency of the incident radiation, which corresponds to the frequency of rotation of the filter wheel 14 shown in Figure 1 multiplied by the number of filters in the filter wheel 14. As can be seen, the detector response is constant for frequencies up to approximately 100 Hz and reaches a point of -3 dB, where signals typically undergo a 90 ° phase shift, at approximately 700 Hz.

Turning now to Figure 7, this represents the detector frequency response for a detector operated in accordance with the present invention. As can be seen, the frequency response of the detector is effectively constant for cutoff frequencies up to about 1000 Hz, and the -3 dB point occurs at about 10 kHz. The frequency response is thus significantly improved.

In practice, the use of the feedback arrangement according to the present invention effectively reduces the magnitude of the signal generated by the detector 282 for amplification by the main amplifier 34 for the output, and this can lead to the introduction of additional noise during amplification. However, by controlling the feedback level with an appropriate selection of the gain of amplifier 38 in the feedback path, the amplitude of the detector's frequency response can be positively selected according to the application.

ES 2 329 754 T3

It is considered that the amplitude of the detector's frequency response will allow measurement times for each measurement reference wavelength as small as 1 ms or less, which is significantly faster than is currently possible.

This decreased response time is particularly advantageous in scanning applications where a moving band of material is scanned to obtain measurement wavelengths, as the results for each scan can be averaged to give excellent cross-sectional resolution of band. The improved response time is also advantageous when the sample whose parameter is to be measured comprises a material, such as snack food and tobacco, susceptible to generating presentation noise, which is due to changes in product height, reflectivity and angle with movement. In this case, fast wavelength data acquisition can allow a significant reduction in presentation noise.

The foregoing description has been limited to the operation of detector 282 and diode 32 to provide an output signal representing light reflected back from sample 22. It will be appreciated that the operation of detector 242 and diode 30 to provide a signal of The output representing the light from the lamp 12 when filtered by the filter wheel 14 is similar. Both signals are then processed in a known way to determine the parameter to be measured.

Various modifications to the described arrangement are possible.

In particular, each of the diodes 30 or 32 can be replaced by an array of said diodes spaced equidistant from the associated detector 242 or 282 to ensure uniform removal of the detector.

Furthermore, diodes 30 and 32 may be arranged to illuminate detectors 24, 28 directly, rather than by reflection from blocking filters 244, 284 as described.

It has been assumed in the above description that detectors 242, 282 are PbS detectors and that diodes 30 and 32 are GaAs diodes. However, other types of detectors can also be used in conjunction with the appropriate diodes.

Furthermore, it will be appreciated that the amplifiers 34, 38 can be replaced by alternative amplifier arrangements as long as the feedback circuit 37 still acts to reduce the current supplied to the diode 30, 32 when the intensity of the light radiating from the detector 242 increases. , 282.