Photodetector amplitude linearity

• US 5,229,838 A
• Filed: 01/21/1992
• Issued: 07/20/1993
• Est. Priority Date: 01/21/1992
• Status: Expired due to Term

• Current Assignee: Perkinelmer Incorporated
• Original Assignee: Perkinelmer Incorporated
• Inventors: Tracy, David H., Ganz, Alan M.
• Primary Examiner(s): EVANS, FANNIE L
• Application Number:
  US07/823,631
• Time in Patent Office: 546 Days
• Field of Search: 356/300, 356/308, 356/319, 356/325, 356/323, 356/326, 356/328, 356/218, 356/222, 356/225, 250/252.1 A
• US Class Current: 356/328
• CPC Class Codes: G01J 1/08 Arrangements of light sources specially adapted for photometry standard sources, also using luminescent or radioactive material
• CPC Class Codes: G01N 21/274 Calibration, base line adjustment, drift correction

Abstract

To calibrate a photodetector, a rotating disk with a slot is disposed in a light beam with decreasing speed from a defined maximum rotational speed to a defined minimum speed, while magnitudes and times of signals are read out and stored. Vernier pairs of signals occur in adjacent readout intervals, and non-vernier signals exclude the verniers. Readout times for verniers are used to estimate a preliminary function of rotations versus time. From the function are estimated an occurrence time for each pair and period of disk rotation at the time. Vernier fraction is the ratio of one signal in the pair to the sum of the pair. A time offset is the product of vernier fraction, slot fraction of the disk and the estimated period. Occurrence times corrected with the time offset are utilized to fit a corrected function of disk rotations versus time. Points of time for the non-vernier signals are determined from the corrected function, each point corresponding to disk rotations to a corresponding non-vernier signal. These points of time are employed with corresponding signals for linearly calibrating the photodetector.

Claims

1. A method of determining linearity of a photodetector having sequential readouts of signals, each signal correlating with amount of light received during a corresponding readout interval, each readout having a time of readout, and all readout intervals being equal to a predetermined readout period, the method comprising:

directing a train of light pulses to the photodetector, the light pulses having equal intensity and a variable pulse period between successive pulses, and each light pulse having a variable pulse width equal to a constant fraction of pulse period;

selecting the constant fraction and a maximum pulse period cooperatively so that each pulse width is not greater than the readout period, and further selecting a minimum pulse period not less than two times the readout period;

smoothly changing the pulse period between the maximum pulse period and the minimum pulse period, while reading out and storing sequential signals from the photodetector and corresponding times of readout;

identifying a set of vernier pairs of signals as those pairs of signals occurring in adjacent readout intervals, and establishing a set of non-vernier signals excluding the vernier pairs;

ascertaining occurrence times for the vernier pairs and a corresponding total number of light pulses to each occurrence time;

utilizing the occurrence times to fit an occurrence function of total number of light pulses versus time; and

correlating the non-vernier signals to the occurrence function so as to effect a correlation function of the non-vernier signals versus pulse period as a measure of linearity of the photodetector.

2. The method of claim 1 wherein the step of correlating comprises determining from the occurrence function point of time for the non-vernier signals, computing from the occurrence function a further function representative of pulse period versus time, identifying from the further function a corresponding pulse period for each point of time and thereby for each corresponding non-vernier signal, and effecting the correlation function with the non-vernier signals versus corresponding pulse period.

3. The method of claim 1 wherein the step of ascertaining occurrence times comprises:

fitting a preliminary function of number of light pulses versus times of readout of firstly occurring signals of the vernier pairs;

computing occurrence times for each vernier pair by determining from the preliminary function a preliminary occurrence time for the pair and a preliminary pulse period at the preliminary occurrence time, defining a vernier fraction as the ratio of one signal in the pair to a sum of signals constituting the pair, and calculating a time offset as a product of the vernier fraction and the constant fraction and the preliminary period; and

correcting the preliminary occurrence time with the time offset, thereby generating a set of corrected occurrence times.

4. The method of claim 3 further comprising iterating at least once the steps of computing occurrence times, by initially utilizing a previously fitted occurrence function as a new preliminary function.

5. The method of claim 1 wherein the step of directing a train of light pulses comprises disposing in a stable light beam a rotating disk having a slot therein with a selected slot width so as to pass the light pulses to the photodetector, and the step of smoothly changing the pulse period comprises rotating the disk at smoothly changing speed.

6. The method of claim 5 wherein the step of rotating is effected by initially driving the disk to greater than the maximum rotational speed, and freeing the disk to coast from the maximum rotational speed to the minimum rotational speed.

7. The method of claim 1 wherein the photodetector is disposed in a spectrophotometer that includes dispersion means receptive of the light pulses so as to direct spectrally dispersed light pulses to the photodetector.

8. The method of claim 7 wherein the photodetector comprises a plurality of photosensitive pixels each receptive of a different spectral portion of the dispersed light pulses, and the method further comprises employing at least two of the pixels for producing the sequential signals.

9. The method of claim 8 wherein the step of utilizing the occurrence times includes correcting said times for differences thereof among the pixels.

10. A precision photometric instrument comprising:

a photodetector having sequential readouts of signals, each signal correlating with amount of light received during a corresponding readout interval, each readout having a time of readout, and all readout intervals being equal to a predetermined readout period;

light means for directing a train of light pulses to the photodetector, the light pulses having equal intensity and a variable pulse period between successive pulses, and each light pulse having a variable pulse width equal to a constant fraction of pulse period;

changing means for smoothly changing the pulse period between a maximum pulse period and a minimum pulse period as to effect a series of sequential signals, the minimum pulse period being not less than two times the readout period, and the maximum pulse period and the constant fraction cooperatively being such that each pulse width is not greater than the readout period;

storing means for storing sequential signals from the photodetector and corresponding times of readout;

means for identifying a set of vernier pairs of signals as those pairs of signals occurring in adjacent readout intervals, and establishing a set of non-vernier signals excluding the vernier pairs;

means for ascertaining occurrence times for the vernier pairs and a corresponding total number of light pulses to each occurrence time;

means for utilizing the occurrence times to fit an occurrence function of total number of light pulses versus time; and

means for correlating the non-vernier signals to the occurrence function so as to effect a correlation function of the non-vernier signals versus pulse period as a measure of linearity of the photodetector.

11. The instrument of claim 10 wherein the means for correlating comprises means for determining from the occurrence function points of time for the non-vernier signals, means for computing from the occurrence function a further function representative of pulse period versus time, means for identifying from the further function a corresponding pulse period for each point of time and thereby for each corresponding non-vernier signal, and means for effecting the correlation function with the non-vernier signals versus corresponding pulse period.

12. The instrument of claim 10 wherein the means for ascertaining occurrence times comprises means for fitting a preliminary function of total number of light pulses versus times of readout of firstly occurring signals of the vernier pairs, means for determining from the preliminary function a preliminary occurrence time for the pair and a pulse period at the preliminary occurrence time, means for defining a vernier fraction as the ratio of one signal in the pair to a sum of signals constituting the pair, means for calculating a time offset as a product of the vernier fraction and the constant fraction and the pulse period, and means for correcting the preliminary occurrence time with the time offset.

13. The instrument of claim 10 wherein the means for directing a train of light pulses comprises a light source for emitting a stable light beam, and a disk disposed in the light beam and having a slot therein with a selected slot width so that upon rotation the disk passes the light pulses to the photodetector, and the means for smoothly changing the pulse period comprises means for rotating the disk at smoothly changing speed.

14. The instrument of claim 13 wherein the means for rotating comprises means for initially driving the disk to greater than the maximum rotational speed, and means for freeing the disk to coast from the maximum rotational speed to the minimum rotational speed.

15. The instrument of 10 wherein the photodetector is disposed in a spectrophotometer that includes dispersion means receptive of the light pulses so as to direct spectrally dispersed light pulses to the photodetector.

16. The instrument of claim 15 wherein the photodetector comprises a plurality of photosensitive pixels each receptive of a different spectral portion of the dispersed light pulses, and the selected signals include signals from at least two of the pixels.

17. The instrument of claim 16 wherein the means for utilizing the occurrence times includes means for correcting said times for differences thereof among the pixels.