Measurement of contents of tanks etc. with microwave radiations

• US 4,044,355 A
• Filed: 02/13/1976
• Issued: 08/23/1977
• Est. Priority Date: 11/20/1973
• Status: Expired due to Term

• Current Assignee: Saab-Scania Aktiebolag (Saab AB)
• Original Assignee: Saab-Scania Aktiebolag (Saab AB)
• Inventors: Edvardsson, Kurt Olov
• Primary Examiner(s): Tubbesing, T.H.
• Application Number:
  US05/658,075
• Time in Patent Office: 557 Days
• Field of Search: 343/12 R, 343/14, 343/17, 343/7
• US Class Current: 342/124
• CPC Class Codes: G01F 23/284 Electromagnetic waves
• CPC Class Codes: G01S 13/343 using sawtooth modulation

Abstract

A microwave signal that varies linearly in frequency through a sweep period is radiated to and reflected back from a target body. In one mixer the directly generated signal is mixed with the reflected signal to produce a difference frequency corresponding to distance; another mixer mixes the directly generated signal with the same signal delayed for a constant time, to produce a reference difference frequency. Each difference frequency is digitized as a pulse train. The sweep period is divided into successive short time intervals, each of which can begin with a pulse of the lower frequency train and to each of which is assigned a numerical weighting factor, said factors for successive intervals differing stepwise in value. Pulses of each train occurring during each such interval are multiplied by the factor for the interval. All such multiplied pulses are counted through the sweep period to obtain a pair of pulse totals, and a quotient relationship between those totals gives an accurate measure of distance.

Claims

1. A method of measuring distance to a target surface by generating a microwave signal that varies in frequency substantially regularly through a predetermined time period, radiating a first part of said generated signal towards the target surface to produce a return signal reflected from the target surface, and mixing said return signal with a second part of the generated signal to obtain an oscillatory output having a difference frequency which is a function of the distance to be measured, said method being characterized by steps which enable a final output to be obtained that accurately denotes the distance travelled by the return signal notwithstanding any irregularities in said oscillatory output that are due to disturbances in the return signal and lack of perfect regularity in the frequency variation of the generated microwave signal, said steps comprising:

A. mixing a third part of said generated signal with a fourth part thereof that has been subjected to a predetermined delay to obtain a second oscillatory output having a reference frequency which corresponds to an arbitrary fixed distance;

B. producing first and second pulsing outputs which correspond respectively to said first mentioned and said second oscillatory outputs and each of which has a pulse rate that is in a predetermined relationship to the frequency of its corresponding oscillatory output;

C. assigning a numerical factor to each of a number of short and substantially equal time intervals, each beginning upon the production of a pulse of one of said pulsing outputs and which together span a major portion of said time period, the factors assigned to successive time intervals differing stepwise in value from one another;

D. during each of said time intervals producing two sets of averaging pulses, one set for each of said pulsing outputs, each set consisting of a number of averaging pulses corresponding to the number of pulses of its pulsing output that are produced during the time interval multiplied by the numerical factor assigned to the time interval;

E. counting the averaging pulses of all of said sets corresponding to each pulsing output that are produced during said major portion of said time period, to obtain a pair of averaging pulse totals; and

F. producing an output which corresponds to a quotient relationship between said averaging pulse totals and which constitutes said final output.

2. The method of claim 1, wherein the issuance of each pulse of said first pulsing output occurs at the beginning of one of said time intervals, further characterized by:

1. so producing said first pulsing output that each pulse thereof corresponds to a half-cycle of said first oscillatory output; and

2. so producing said second pulsing output that the pulse rate thereof is higher than the frequency of said second oscillatory output and substantially higher than the pulse rate of said first pulsing output.

3. The method of claim 2 wherein said second pulsing output is produced by:

1. multiplying the frequency of said second oscillatory output by a constant value to produce a modified second oscillatory output having a frequency which corresponds to a multiple of said arbitrary fixed distance; and

2. producing a pulse of said second pulsing output at each half-cycle of said modified oscillatory output.

4. The method of claim 3 wherein the frequency of said second oscillatory output is multiplied by a constant value so selected that the frequency of the modified second oscillatory output is higher than the highest expectable value of the frequency of the first oscillatory output.

5. The method of claim 1 wherein:

said numerical factor that is assigned to each of said time intervals is equal to n(N - n), where

N is the number of said time intervals in said major portion of said time period, and

n is the ordinal number of the particular time interval during said time period.

6. The method of claim 1, further characterized by:

G. before producing said first pulsing output, filtering said first oscillatory output by passing the same through a narrow-band bandpass filter having a controllably variable center frequency; and

H. so controlling the center frequency of said filter that the same is at all times related to

a. the approximate distance to be measured and

b. the frequency of said second oscillatory output.

7. The method of claim 1, further characterized by:

1. comparing the amplitude of the first oscillatory output with a predetermined reference level, and

2. discarding oscillatory outputs obtained for any time period during which said amplitude is not above said reference level.

8. A method of measuring distance to a target surface by generating a microwave signal that varies in frequency substantially regularly through a predetermined time period, radiating a first part of said generated signal towards the target surface to produce a return signal reflected from the target surface, and mixing said return signal with a second part of the generated signal to obtain an oscillatory output having a difference frequency which is a function of the distance to be measured, said method being characterized by steps which enable a final output to be obtaned that accurately denotes the distance travelled by the return signal notwithstanding any irregularities in said oscillatory output that are due to disturbances in the return signal and lack of perfect regularity in the frequency variation of the generated microwave signal, said steps comprising:

A. producing a second oscillatory output having a reference frequency which is substantially higher than that of the first oscillatory output by

1. mixing a third part of said generated signal with a fourth part thereof that has been subjected to a predetermined delay and

2. multiplying by a fixed constant value the difference frequency signal obtained as a result of the last mentioned mixing;

B. producing

9. a first pulsing output that provides a pulse for every half-cycle of said first oscillatory output, anda second pulsing output that provides a pulse for every half cycle of said second oscillatory output;

C. assigning a numerical weighting factor to each of the successive short time intervals that begins with the issuance of each pulse of said first pulsing output during a major portion of said time period, which weighting factor is equal to n(N-n), where

N denotes the number of said time intervals in said portion of said time period, and

n denotes the ordinal number of the particular time interval in said time period;

D. multiplying each pulse of the first pulsing output issued during said portion of said time period by the value of n(N-n) for the time interval in which such pulse is issued;

E. storing the products of all such multiplications during the time period to obtain a first weighted pulse count;

F. multiplying each pulse of said second pulsing output by the value of n(N-n) for the time interval in which that pulse is issued;

G. storing the products of all of the last mentioned multiplications during said time period to obtain a second weighted pulse count; and

H. at the conclusion of said time period producing an output which corresponds to a quotient relationship between said first and second weighted pulse counts and which consitutees said final output.