Weight measuring method using a plurality of sensors

• US 5,805,467 A
• Filed: 07/11/1997
• Issued: 09/08/1998
• Est. Priority Date: 07/02/1996
• Status: Expired due to Fees

• Current Assignee: James L. Richards
• Original Assignee: James L. Richards
• Inventors: Richards, James L.
• Primary Examiner(s): Trammell, James P.
• Assistant Examiner(s): MILLER, CRAIG S
• Application Number:
  US08/893,406
• Time in Patent Office: 424 Days
• Field of Search: 364/550, 364/570, 364/571.01, 364/567, 177/164, 177/25.11, 177/25.12, 177/25.13, 177/25.14, 119/842, 73/1 B
• US Class Current: 702/101
• CPC Class Codes: G01G 23/01 Testing or calibrating of weighing apparatus
• CPC Class Codes: G01G 23/3707 using a microprocessor
• CPC Class Codes: G01G 3/1416 for non-linearity

Abstract

A weight measuring apparatus comprising a plurality of load sensors, together with an input network, a multiplexer, a reference, an analog to digital converter, and a processor, wherein the input network includes an analog summing network and direct sensor connections to the multiplexer. The multiplexer is arranged to permit measurement of the sum signal, a reference signal, and each direct load sensor signal in a prescribed sequence. The multiplexed signals are converted to digital and processed to derive a weight estimate based on the multiplexed measurements and a system error model. The prescribed measurement sequence permits measurement of the sum signal more frequently than each of the direct signals. A method is described for deriving correction factors for the error model comprising the placing of calibration weights primarily on each weight sensor and solving the system of simultaneous equations for the resulting correction factors.

Claims

1. A method of increasing the output rate of a scale system which multiplexes among a plurality of sensors comprising the following steps:

(1) producing a set of sensor signals from a plurality of load sensors;

(2) generating an approximate weight estimate signal based at least in part on the sensor signals;

(3) sampling the approximate weight signal and the sensor signals according to a prescribed sampling sequence wherein the approximate weight signal is sampled more frequently than at least one of the sensor signals;

(4) converting the sampled signals to digital representations;

(5) processing the digital representations of the sensor signals and the approximate weight signal together with a system error model to derive a corrected weight value;

(6) outputting the corrected weight in association with the sampling of the approximate weight signal.

2. Method as in claim 1 further including deriving the approximate weight estimate signal of step (2) from a resistor network connected to the sensor signal outputs.

3. Method as in claim 1 wherein the scale includes a circuit for generating a reference signal and the sampling step (3) includes sampling the reference signal in association with each sensor signal and each approximate weight signal, and the digitized reference measurements are included in the processing step (5) to derive the corrected weight value.

4. Method as in claim 1 further including deriving the reference signal of step (2) from a resistor network connected to the sensor network.

5. Method of reducing the error in a scale system which multiplexes among a plurality of signals comprising the following steps:

(1) producing at least one sensor signal;

(2) producing a reference signal which is stable in relationship to the sensor signal such that the difference between the reference signal and the sensor signal is useful in determining the load weight;

(3) sampling the reference signal and the sensor signals according to a prescribed sampling sequence wherein the reference signal is sampled immediately before or immediately after a sensor signal;

(4) converting the sampled signals to digital representations;

(5) processing the digital representations in accordance with a system error model to derive a corrected weight value.

6. Method as claimed in claim 5 wherein the sampling step (3) includes sampling the reference signal both before and after a sensor signal and both reference measurements are utilized together with the sensor signal as part of the determination of the load weight.

7. Method as claimed in claim 6 wherein step (4) includes averaging the two reference signals to produce a derived reference signal to be used with the sensor signal as part of the determination of the load weight.

8. Method as claimed in claim 5 wherein at least one of the sampled signals is an approximate sum weight estimate signal derived at least in part from the sensor signals.

9. Method as in claim 5 further including deriving the reference signal of step (2) from a resistor network connected to the sensor network.

10. A method of compensating a weight measuring scale comprising the following steps:

(1) producing a set of sensor signals from a plurality of load sensors;

(2) generating an approximate weight estimate signal based, at least in part, on the sensor signals;

(3) determining a mathematical expression for corrected weight based at least in part on the generated weight estimate signal and the sensor signals; and

(4) deriving a corrected weight measurement based at least in part on the determined expression for corrected weight, the sensor signals and the weight estimate signal.

11. Method as in claim 10 wherein the determining step (3) includes determining at least one constant for a particular weight measuring scale, and storing the constant.

12. Method as in claim 11 wherein the constant determining step (3) includes repetitively coupling known loads to the load sensors in a way such that each successive sensor responds greater than any of the others, recording the sensor signals, and using the recorded signals to, at least in part, determine the constant.

13. Method as in claim 12 wherein the constants are adjusted such that one constant associated with a load sensor is zero.

14. Method as in claim 10 further including deriving the approximate weight estimate signal of step (2) from a resistor network connected to the sensor outputs.

15. Method as in claim 10 wherein the sampling step (3) includes sampling the approximate weight signal according to a prescribed sampling sequence.

16. Method as in claim 10 wherein the scale includes a circuit for generating a reference signal;

the sampling step (3) includes sampling the reference signal in association with each sensor signal and each approximate weight signal; and

the digitized reference measurements are included in the processing step (5) to derive the corrected weight value.

17. Method as in claim 10 wherein the sampling step (3) includes sampling the approximate weight signal according to a prescribed sampling sequence.

18. Method as in claim 10 further including deriving the approximate weight estimate signal of step (2) from a resistor network connected to the sensor signal outputs.