Apparatus and methods for performing acoustical measurements

• US 6,279,379 B1
• Filed: 11/19/1999
• Issued: 08/28/2001
• Est. Priority Date: 04/17/1998
• Status: Expired due to Term

• Current Assignee: Veeco Instruments Incorporated
• Original Assignee: Loren Industries Inc.
• Inventors: Lee, Patrick S., Sirota, Don N., Logue, Raymond C.
• Primary Examiner(s): Larkin, Daniel S.
• Assistant Examiner(s): WIGGINS, JOHN DAVID
• Application Number:
  US09/443,519
• Time in Patent Office: 648 Days
• Field of Search: 73/24.01, 73/23.2, 73/597, 73/579, 73/602, 324/76.21, 324/76.33
• US Class Current: 73/24.01
• CPC Class Codes: G01N 2291/02809 Concentration of a compound, e.g. measured by a surface mass change
• CPC Class Codes: G01N 2291/102 one emitter, one receiver
• CPC Class Codes: G01N 29/024 by measuring propagation velocity or propagation time of acoustic waves
• CPC Class Codes: G01N 29/222 Constructional or flow details for analysing fluids optoacoustic fluid cells G01N29/2425
• CPC Class Codes: G01N 29/449 Statistical methods not provided for in G01N29/4409, e.g. averaging, smoothing and interpolation
• CPC Class Codes: G01N 29/46 by spectral analysis, e.g. Fourier analysis or wavelet analysis spectral signal processing per se G06F17/14
• CPC Class Codes: G01N 29/50 using auto-correlation techniques or cross-correlation techniques

Abstract

Apparatus (15, 30) and methods for performing acoustical measurements are provided having some and preferably all of the following features: (A) the system (15, 30) is operated under near-field conditions; (B) the piezoelement (40) or piezoelements (40, 48) used in the system are (i) mechanically (41, 49) and electrically (13, 16) damped and (ii) efficiently electrically coupled to the signal processing components of the system; (C) each piezoelement (40, 48) used in the system includes an acoustical transformer (42, 50) for coupling the element to a gaseous test medium (9); (D) speed of sound is determined from the time difference between two detections of an acoustical pulse (81, 82) at a receiver (40, FIG. 3; 48, FIG. 7); (E) cross-correlation techniques are employed to detect the acoustical pulse at the receiver; (F) forward and inverse Fourier transforms employing fast Fourier transform techniques are used to implement the cross-correlation techniques; in such a mathematical manner that the peak of the cross-correlation function corresponds to the detection of a pulse at the receiver and (G) stray path signals through the body (31) of the acoustic sensor (15, 30) are removed from detected signals prior to signal analysis. Techniques are also provided for performing acoustical measurements on gases whose thermodynamic properties have not been measured and on mixtures of compressible gases. Methods and apparatus (29) for performing feedback control of a gas of interest in a mixture of that gas and a carrier gas are provided in which the controlled variable is the flow of the carrier gas.

Claims

1. Apparatus for use with (i) a transmitter which produces a pulse of acoustical energy (the transmitted pulse) which passes through a gas, liquid, or solid, and (ii) a receiver which detects the transmitted pulse after its passage through the gas, liquid, or solid (the detected pulse), said apparatus comprising:

(a) cross-correlating means associated with the receiver for performing a cross-correlation between (i) a stored calibration waveform representing the transmitted pulse and (ii) the detected pulse, said cross-correlating means producing a cross-correlation function for the detected pulse; and

(b) peak determining means for determining the peak of said cross-correlation function, said peak being indicative of the time the detected pulse was detected by the receiver;

wherein the stored calibration waveform is not directly derived from the transmitted pulse but is determined during calibration of the apparatus and stored prior to the production of said transmitted pulse by the transmitter.

2. The apparatus of claim 1 wherein the transmitted pulse comprises acoustical oscillations having a center frequency and the stored calibration waveform is determined using a gas which has:

(i) either a known specific heat, compressibility, and molecular weight, or a known speed of sound, and (ii) a close to ideal gas behavior at the center frequency.

3. The apparatus of claim 2 wherein the stored calibration waveform is determined by averaging a large number of detected and stored pulses.

4. The apparatus of claim 1 wherein the cross-correlation is performed by:

(i) performing a fast Fourier transform of the detected pulse to produce a Fourier transformed detected pulse;

(ii) multiplying the Fourier transformed detected pulse by the complex conjugate of a Fourier transform of the stored calibration waveform to produce a Fourier space cross-correlation function; and

(iii) performing an inverse fast Fourier transform of said Fourier space cross-correlation function and taking the real part of said inverse fast Fourier transform to produce said cross-correlation function.

5. The apparatus of claim 1 wherein the peak of the cross-correlation function is determined by:

(i) identifying the largest amplitude of the cross-correlation function and two amplitudes adjacent to the largest amplitude;

(ii) fitting a second order polynomial to said three amplitudes; and

(iii) determining the peak of the cross-correlation function from said second order polynomial.

6. The apparatus of claim 2 wherein the transmitted pulse has leading and trailing cycles and is windowed so that said leading and trailing cycles taper from zero to full amplitude.

7. The apparatus of claim 2 wherein the transmitted pulse is frequency modulated.

8. The apparatus of claim 2 wherein the transmitter and receiver are separated by a distance which defines a wraparound time and the transmitted pulse has a duration which is longer than the wraparound time.

9. A method for testing a gas, liquid, or solid comprising:

(a) producing a pulse of acoustical energy which passes through the gas, solid, or liquid (the transmitted pulse);

(b) detecting the transmitted pulse after it has passed through the gas, solid, or liquid (the detected pulse);

(c) performing a cross-correlation between (i) a stored calibration waveform representing the transmitted pulse and (ii) the detected pulse, said cross-correlation producing a cross-correlation function for the detected pulse; and

(d) determining the peak of said cross-correlation function, said peak being indicative of the time the detected pulse was detected in step (b);

wherein the stored calibration waveform is not directly derived from the transmitted pulse but is determined during a calibration procedure and stored prior to the production of said transmitted pulse in step (a).

10. The method of claim 9 wherein the transmitted pulse comprises acoustical oscillations having a center frequency and the stored calibration waveform is determined using a gas which has:

(i) either a known specific heat, compressibility, and molecular weight, or a known speed of sound, and (ii) a close to ideal gas behavior at the center frequency.

11. The method of claim 10 wherein the stored calibration waveform is determined by averaging a large number of detected and stored pulses.

12. The method of claim 9 wherein the cross-correlation is performed by:

(i) performing a fast Fourier transform of the detected pulse to produce a Fourier transformed detected pulse;

(ii) multiplying the Fourier transformed detected pulse by the complex conjugate of a Fourier transform of the stored calibration waveform to produce a Fourier space cross-correlation function; and

(iii) performing an inverse fast Fourier transform of said Fourier space cross-correlation function and taking the real part of said inverse fast Fourier transform to produce said cross-correlation function.

13. The method of claim 9 wherein the peak of the cross-correlation function is determined by:

(i) identifying the largest amplitude of the cross-correlation function and two amplitudes adjacent to the largest amplitude;

(ii) fitting a second order polynomial to said three amplitudes; and

(iii) determining the peak of the cross-correlation function from said second order polynomial.

14. The method of claim 9 wherein the transmitted pulse has leading and trailing cycles and is windowed so that said leading and trailing cycles taper from zero to full amplitude.

15. The method of claim 9 wherein the transmitted pulse is frequency modulated.

16. The method of claim 9 wherein steps (a) and (b) are performed at locations which define a wraparound time and the transmitted pulse has a duration which is longer than the wraparound time.

17. Apparatus for use with (i) a transmitter which produces a pulse of acoustical energy (the original pulse) which passes through a gas, liquid, or solid, and (ii) a receiver which detects the pulse after its passage through the gas, liquid, or solid (the detected pulse), said apparatus comprising:

(a) cross-correlating means associated with the receiver for performing a cross-correlation between a waveform representing the original pulse and the detected pulse, said cross-correlating means producing a cross-correlation function for the detected pulse; and

(b) peak determining means for determining the peak of said cross-correlation function, said peak being indicative of the time the detected pulse was detected by the receiver;

wherein the peak of the cross-correlation function is determined by;

(i) identifying the largest amplitude of the cross-correlation function and two amplitudes adjacent to the largest amplitude;

(ii) fitting a second order polynomial to said three amplitudes; and

(iii) determining the peak of the cross-correlation function from said second order polynomial.

18. Apparatus for use with (i) a transmitter which produces a pulse of acoustical energy (the original pulse) which passes through a gas, liquid, or solid, and (ii) a receiver which detects the pulse after its passage through the gas, liquid, or solid (the detected pulse), said apparatus comprising:

(a) cross-correlating means associated with the receiver for performing a cross-correlation between a waveform representing the original pulse and the detected pulse, said cross-correlating means producing a cross-correlation function for the detected pulse; and

(b) peak determining means for determining the peak of said cross-correlation function, said peak being indicative of the time the detected pulse was detected by the receiver;

wherein the transmitter and receiver are separated by a distance which defines a wraparound time and the original pulse has a duration which is longer than the wraparound time.

19. A method for testing a gas, liquid, or solid comprising:

(a) producing a pulse of acoustical energy which passes through the gas, solid, or liquid (the original pulse);

(b) detecting the pulse after it has passed through the gas, solid, or liquid (the detected pulse);

(c) performing a cross-correlation between a waveform representing the original pulse and the detected pulse, said cross-correlation producing a cross-correlation function for the detected pulse; and

(d) determining the peak of said cross-correlation function, said peak being indicative of the time the detected pulse was detected in step (b);

wherein the peak of the cross-correlation function is determined by;

(i) identifying the largest amplitude of the cross-correlation function and two amplitudes adjacent to the largest amplitude;

(ii) fitting a second order polynomial to said three amplitudes; and

(iii) determining the peak of the cross-correlation function from said second order polynomial.

20. A method for testing a gas, liquid, or solid comprising:

(a) producing a pulse of acoustical energy which passes through the gas, solid, or liquid (the original pulse);

(b) detecting the pulse after it has passed through the gas, solid, or liquid (the detected pulse);

(c) performing a cross-correlation between a waveform representing the original pulse and the detected pulse, said cross-correlation producing a cross-correlation function for the detected pulse; and

(d) determining the peak of said cross-correlation function, said peak being indicative of the time the detected pulse was detected in step (b);

wherein steps (a) and (b) are performed at locations which define a wraparound time and the original pulse has a duration which is longer than the wraparound time.