Abstract:
A fast and effective way to step a relative position quantity by a reference interval. Each step in relative position includes open-loop and closed-loop control intervals. The invention contemplates an actuator capable of changing the relative position and a closed loop servo that acts on the actuator to keep the relative position centered on the nearest one of a series of reference values separated by the reference interval. The actuator is preferably capable of a fast response. Stepping the relative position is accomplished as follows, assuming an initial condition where the servo has locked the relative position to a particular initial reference value. First, the actuator is caused to change the relative position by an amount approximately equal to the reference interval in a manner that the servo cannot track the change, such as by disabling the servo. Servo control is then re-established, at which point the servo operates to keep the relative position centered on the nearest reference value. Servo control is maintained until a new step is required, at which time the process is repeated.
Abstract:
An interferometer bearing assembly including a fixed glass rail, a movable carriage that carries both the movable interferometer mirror and the drive coil for the linear motor, and a mechanism for biasing the carriage onto the rail. The glass rail has two optically flat faces defining a normally horizontal line of intersection. The carriage is a rigid structure with plastic surface portions that provide a bearing interface for contacting the flat faces of the glass rail. To improve stability and reduce the possibility of any transverse movement, the carriage is downward biased, preferably by providing an overlying glass plate with a downwardly-facing flat surface, and mounting a spring-loaded plastic element to the top of the carriage. The plastic element contacts (and slides along) the flat glass surface and transmits a downward force to the carriage.
Abstract:
Apparatus and method for calibrating an adaptive ranging A/D converter wherein a noise having known statistical properties is deliberately superposed on the reference DC signals used to calibrate the gain-ranging amplifier. The amplitude of the noise is chosen to be greater than at least one, and preferably several, least significant digits of the A/D converter at the lowest anticipated gain, yet preferably less than half the counting range of the converter at the highest gain. The reference DC signal levels are chosen to be spaced from the minimum and maximum count of the A/D at the lowest and highest anticipated gains by better than half the noise amplitude. A time average of a series of measurements of the noisy reference signals as amplified and digitized by the system are obtained. The statistical properties of the noise generator permits the system output corresponding to the reference signals to be determined to a greater accuracy than the resolution of the A/D, thereby allowing the gain and offset of the gain-ranging amplifier to be determined with greater precision than is possible by injecting noiseless DC signals alone.
Abstract:
A source for reducing drift of a hot spot by producing infrared (IR) light with an insulating housing; an electrical power supply; first and second resistance elements having first and second resistivities; a third resistance element having a third resistivity; the third resistance element disposed between the first and second resistance elements, with the resistivity of the third resistance element having a value greater than the resistivity of both the first and second resistance elements. Alternatively, the first and second resistance elements can have cross-sectional areas that are greater than that of the third resistance. A method of reducing drift of a hot spot in a radiating element of a source of IR illumination by applying constant polarity substantially constant direct current from the power supply and reversing polarity in less than 24 hours, by a square wave AC source, or a Class E amplifier at constant power.
Abstract:
Method and apparatus for providing an imaging attenuated total reflection (ATR) spectrometer which provides faster measurement speed and better spatial resolution than systems collecting an equivalent amount of data using conventional, non-imaging ATR methods and systems. Apparatus includes a radiation source, an interferometer coupled to the radiation source which produces a spectrally-multiplexed input beam of radiation, an internal reflection element (IRE) engaging a sample-under-test, a focal plane array detector, a first optical system adapted and positioned to direct and concentrate the input beam through the rear surface of the IRE towards a contact area of the sample such that an angle of incidence of said input beam at the front surface is equal to or greater than the critical angle for the IRE, and a second optical system adapted and positioned to collect reflected radiation from the contact area and image the same onto the focal plane array detector.
Abstract:
A technique for extracting the impulse response of a sample of interest includes corresponding measurements made with the sample of interest and a reference sample. At each of a series of steps in an FT-IR spectrometer, the sample of interest is illuminated with an excitation pulse of infrared radiation, acoustic signals having a time dependence o.sub.S (t) arising from the excitation pulse are captured, and a Fourier transform O.sub.S of o.sub.S (t) is computed. At each of a series of steps in the FT-IR spectrometer, the reference sample is illuminated with an excitation pulse of analytic radiation, acoustic signals having a time dependence O.sub.R (t) arising from the excitation pulse are captured, and a Fourier transform O.sub.R of o.sub.R (t) is computed. For each step, an inverse Fourier transform of the ratio O.sub.S /O.sub.R is computed to provide a series of values s(t.sub.i) for a series of times t.sub.i. These values s(t.sub.i) represent the impulse response s(t) of the sample of interest for the mix of optical frequencies for that retardation value. Interferograms are processed to provide photoacoustic spectra.
Abstract:
Digital signal proceessing (DSP) techiques for performing multiple modulation measurements with a polarization photoelastic modulator (PEM) in a step-scanning FT-IR spectrometer. The frequency and phase of the PEM drive signal are extracted from the digitized data collected for the actual measurement. This can then be used to perform the desired analysis of the polarization signals (e.g., CD,LD, DIRLD). This is accomplished by successively refining an initial estimate of the PEM frequency (typically starting at the nominal PEM frequency .omega..sub.0, or at the value determined from the previous step). This is done by using the current estimate of the PEM frequency to compute a phase error, and then using the computed phase error to refine the estimate of the PEM frequency. The phase errors are computed using different sets of samples in the sampling interval.
Abstract:
For a step-scanning Fourier transform spectrometer comprising an interferometer, a detector, and a digital signal processor (DSP), a method for measuring a response of a sample to multiple modulations is disclosed. In one embodiment, the method comprises the steps of utilizing the DSP to measure a phase modulation angle and a sample modulation angle; obtain an interferogram that corresponds to a calibrated static sample response by using the measured phase modulation rotation angle; obtain another interferogram which corresponds to the calibrated dynamic sample response by using both the measured phase and sample modulation rotation angles; and computing one or more spectra from the interferograms which indicates the calibrated response of said sample to the multiple modulations.
Abstract:
The present invention provides a technique for improving the resolution of an A/D converter. The input analog signal is sampled to generate an analog level and the analog level is held for an interval. A random noise signal having zero average value is superimposed on the held level to generate a fluctuating voltage. This fluctuating voltage is then sampled a plurality of at least N times, and N sampled values are communicated to the A/D converter so that N digitized values are generated. These digitized values are averaged to provide an output having a digitization error reduced by a factor of N.sup.1/2.