Radiant energy – Photocells; circuits and apparatus
Reexamination Certificate
2000-12-18
2003-08-19
Pyo, Kevin (Department: 2878)
Radiant energy
Photocells; circuits and apparatus
C250S228000, C250S216000
Reexamination Certificate
active
06608293
ABSTRACT:
The inventions concerns a device and method to provide quality control for a photosensor, especially a photodiode array or a photodiode matrix, whose output signal depends on the intensity of an input signal formed by electromagnetic waves, whereby the photosensor to be tested receives stimulation signals forming the input signals while the stimulation signal intensity of the stimulation signals is varied, and whereby the associated output signals of the photosensor to be tested are measured and recorded for evaluation purposes.
Photosensors, especially photodiode arrays, are for example used in spectrometers, color measuring systems, scanners and pattern recognition systems. A photodiode array is formed by a linear arrangement of photosensitive elements, and it generates output signals that are a measure of the light intensity received by the individual photosensitive elements. Numerous different types of such photodiode arrays are obtainable. Simple types consist of just one arrangement of a number of photodiodes with either conventional anodes or cathodes, whereby the individual photodiodes are connected by separate terminal posts for the output signals. To improve the signal precision and/or improve the usefulness or user suitability, the photodiode arrays on chips or wafers have been developed into integrated photodiode arrays with greater functionality. Of particular interest are “active pixel sensors” (APS). Their particular feature is that increasingly complex circuit elements are assigned to each pixel (i.e., each photosensor element). Such active circuit elements can have functions that range from amplification to digitalizing circuit functions. Two main types are differentiated: multiplex photodiode arrays and parallel photodiode arrays.
With multiplex photodiode arrays, specific electronic circuits are integrated on the chip that permit the photosignals to be read out and processed via a single signal path. Depending on the functionality created on the chip, the output signals are either analog or digital. The charges generated by the individual photodiodes are stored between the readout cycles in the connected capacitors via an integration interval. At the end of this integration interval, pixels are read out when the respective charges are transmitted sequentially to a common line via electronic switches. In this case, an A/D converter in the additional signal path is foreseen that sequentially converts the signals into digital values. Frequently, such A/D converters are already included in the silicon chip. Usually “successive approximation”-type converters are used, since they represent a favorable compromise between speed and complexity. Unfortunately with this type of converter, procedural variations frequently lead to differential non-linearity.
With parallel photodiode arrays, analog/digital converters for each photoelement are on the chips. The parallel photodiode arrays provide simultaneously generated digital output signals that are accessible either via serial or parallel bus systems. Thus such a sensor offers simultaneous operation which is very important in many applications. However, the complexity of this sensor is greater, and the frequency of defects accordingly increases during the manufacturing process.
In addition to photodiode arrays with a linear arrangement of photosensors, prior-art matrix photodiode arrays also exist. These have a flat structure, and the pixels are in rows and columns. Frequently, the sensors are used as camera chips or image sensors.
Increasing the functionality on the photosensor chips has made their design, handling and function increasingly complex. This also increases the probability of defects during the manufacturing process. Specialized tests must therefore be carried out to check the properties of such photosensors during the manufacturing process for quality control. This requires highly specialized optical stimulation signals and highly specialized test methods that satisfy the requirements for precision and test time. Typical photosensor properties to be tested and characterized are the input/output function concern integral and differential linearity, the response function or output signals as a function of the wave lengths, the noise or disturbance intensity as a function of the signal intensity of the input signals, and the uniformity or homogeneity of the response signals of each photoelement in the photodiode array. In addition, the mechanical limitations need to be considered, especially the accessibility of the photosensors during tests on the wafer level carried out during manufacture.
The input/output transmission function must be tested to provide quality control of such photosensors. An optical stimulus signal is generated that is applied to the sensor whose response or output signal is measured with varied input signal intensity. The respective stimulation signal intensity must be known in this context. This is determined with the aid of a reference sensor whose input/output transmission function is known. Normally a measurement is first made using the reference sensor that is mounted at the device position at which the photosensor to be tested will later be mounted.
As a first step, it is therefore necessary to measure and characterize the stimulation signal using the reference sensor. Then in a second step, the reference sensor must be removed and replaced with the photosensor to be tested (DUT). Then another measurement must be made using the same stimulation signal in regard to its time characteristic and intensity under the same test and environmental conditions. By measuring and recording the respective output signals for evaluation purposes, the results of both measurements can be compared which serves as a measure for the quality of the photosensor to be tested. It is necessary to make the two measurements (of the setpoints using the reference sensor and of the actual values using the photosensor to be tested (DUT)) one right after the other or, in any case, separated only by a brief interval. Otherwise, substantial uncertainty can sometimes arise from the intermediate change in environmental conditions or changed radiation pattern of the light source, for example due to aging over time. The calibration measurement therefore will also have to be repeated at certain intervals using the reference sensor, and the reference sensor and sensor to be tested (DUT) will have to be mounted and removed each time.
This conventional state of the art procedure has a series of problems and difficulties. Changing the sensor once or several times causes uncertainty from altering the precise three-dimensional position of the sensor that can produce corresponding measurement uncertainty or imprecision. Further uncertainty can arise from differences in the respective types of photosensors, i.e., the reference sensor and the sensors to be tested, and in the properties of the output signals and the measuring setup. The previously required installation and removal procedure leads to long test cycles and allows little or no automation in quality control.
In particular when measuring small changes in the stimulation signal intensity, for example in photometric measuring procedures using spectrometers, absorbance detectors, color measuring systems or in the case of absorption measurements, faults in differential linearity of the photosensors have a disturbing influence and are characterized by the differential linearity (DNL) of the photosensor to be tested. Such faults distort the measuring results and increase the quantization noise. DNL has been determined to date by making a small continuous change in the input signal and measuring the associated output signals.
In the case of photosensors with digital output signals, DNL is defined as the deviation of the individual quantization steps from the average expected quantization step. This DNL can arise in the form of “missing codes” (FIG.
3
), “double codes” (FIG.
4
), or as a “dead zone” (FIG.
5
). In contrast,
FIG. 6
shows a fault-free input/output signal tra
Agilent Technologie,s Inc.
Pyo Kevin
LandOfFree
Method of and apparatus for testing a photosensor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of and apparatus for testing a photosensor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of and apparatus for testing a photosensor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3114121