Method of and apparatus for ascertaining the fine position value of a movable body

Patent metadata

Field	Value
Publication number	US-9587962-B2
Application number	US-201113501167-A
Country	US
Kind code	B2
Filing date	Feb 23, 2011
Priority date	Mar 5, 2010
Publication date	Mar 7, 2017
Grant date	Mar 7, 2017

Abstract

A position sensor for ascertaining the fine position value z of a movable body includes an exciter unit moving therewith and a stationary sensor unit ( 7 ) which simultaneously delivers a plurality of output signals jointly describing the fine position value. In a calibration mode there is established a defined relationship between the exciter unit and a calibration unit ( 31 ) such that groups of amplitude values can be taken off from said output signals and groups of average values are formed therefrom, which are fed to the calibration unit ( 31 ) which converts them into reference values using the calibration fine position values μ(z) and stores same with the associated fine position as an associated values multiplet in a comparative value memory ( 14 ). In each measuring mode groups of amplitude are taken off from the output signals from which groups of measuring values are produced by averaging and are fed to a computing unit ( 10 ), which forms differences of cross products from the reference values of varying groups of reference values and the current measuring value group and causes said differences to go towards zero to ascertain the current fine position value.

First claim

The invention claimed is:

1. A method of ascertaining a fine position value describing the instantaneous absolute position of a movable body, said method comprising the step of providing a position sensor which comprises:
an exciter unit which moves with said movable body and which can pass through a predetermined range of movement, said range comprising at least one measuring segment;
a stationary sensor unit ( 7 ) adapted to generate a plurality of output signals which jointly describe the variation in time of said fine position value;
signal processing means adapted to receive said output signals and, at predeterminable points of time, to derive from said output signals average values jointly describing said fine position value of said body at a respective point of time,
wherein said position sensor further comprises:
a comparative value memory ( 14 ) adapted to store reference values;
a computing unit ( 10 ) adapted to compare said average values with said reference values;
a calibration unit ( 31 ) adapted to receive said average values; and
a regulating loop consisting of said computing unit ( 10 ), said comparative value memory ( 14 ) and a feedback circuit ( 15 ), and

wherein said method further comprises the following steps:
performing at least one calibration run for calibrating said position sensor, said calibration run comprising the sub-steps of
additionally providing a calibration standard ( 30 ) comprising
a movable scale which also moves with said movable body and, being in a defined relationship with the exciter unit of the position sensor, passes through a corresponding predetermined range of movement; and
a stationary detector adapted to sense said movable scale and to generate a calibration fine position value signal representing the variation in time of the instantaneous position of said movable body with high accuracy;

sampling said calibration fine position value signal at predeterminable points of time to obtain calibration fine position values representing said fine position value of said body at the respective point of time and, at each of said points of time, simultaneously sampling said output signals delivered by said stationary sensor unit ( 7 ) in order to obtain associated average reference values
feeding each of said calibration fine position values and the associated average reference values to said calibration unit ( 31 ), which puts them together to form multiplets of associated values having a defined relationship and stores said multiplets in said comparative value memory ( 14 ) in a retrievable manner;
operating after said calibration run said position sensor without said calibration standard ( 30 ) in a measuring mode for ascertaining the current fine position value by the following sub-steps:
feeding the output signals of sensor unit ( 7 ) to said signal processing means which derive from said output signals average measurement values;
feeding said average measurement values to said computing unit ( 10 );
causing said comparative value memory ( 14 ) under the control of said feedback circuit ( 15 ) to feed a set of average reference values to said computing unit ( 10 ) which compares said average reference values with said average measurement values by forming cross products for each set of average reference values and determines the magnitude of the differences of such cross products;
repeating that last sub-step under the control of said feedback circuit ( 15 ) until said differences have reached a value closest to zero; and
identifying the associated fine position value as being the fine position value of the instantaneous absolute position of said movable body.

2. A method as set forth in claim 1 wherein said method further comprises the step of providing signal processing means being adapted to derive average values from said output signals by the following sub-steps:
sampling each of said output signals several times at points of time using a high scanning clock frequency such that the position of the body to be monitored does not markedly change during said sampling procedure to obtain groups of individual amplitude values;
forming a first average value from a predetermined number of previously ascertained individual amplitude values;
providing said first average value with a first weighting factor and providing the latest individual value with a second weighting factor; and
forming a new average value from those two weighted values.

3. A method as set forth in claim 1 wherein the step of providing a position sensor comprises
providing a stationary sensor unit ( 7 ) delivering two output signals from which by said signal processing means two average values are derived so that, during the calibration step, triplets of associated values are generated an stored in said comparative value memory ( 14 ) each triplet comprising one calibration fine position value and two associated average reference values; and
wherein, in the measuring mode, the comparison of said two average reference values and said two average measurement values performed by said computing unit ( 10 ) comprises the sub-steps of
forming cross products and their difference value; and
feeding that difference value to said feedback circuit ( 15 ) which checks whether said difference value is larger than, smaller than or equal to 0 and derives in dependence of the result of that check a control signal which is fed to a control input of the comparative value memory ( 14 ), said control signal causing the comparative value memory ( 14 ) to send different sets of average reference values to said computing unit ( 10 ) until the condition of the difference value equal to 0 is met, which indicates that the difference has reached a value which is closest to zero.

4. A method as set forth in claim 3 wherein said method further comprises the step of providing signal processing means being adapted to derive average values from said output signals by the following sub-steps:
sampling each of said output signals, several times at points of time using a high scanning clock frequency such that the position of the body to be monitored does not markedly change during said sampling procedure to obtain groups of individual amplitude values
forming a first average value from a predetermined number of previously ascertained individual amplitude values;
providing said first average value with a first weighting factor and providing the latest individual value with a second weighting factor; and
forming a new average value from those two weighted values.

5. A method of ascertaining the position of a body to be monitored, the range of movement of which is subdivided into a plurality of measuring segments which on the one hand are absolutely counted and for which on the other hand fine position values are ascertained as set forth in claim 1 , wherein in the calibration run as well as in the measuring mode the output signals of the sensor unit ( 7 ) are respectively at least semi-periodic, wherein the semi-period length respectively corresponds to the length of the associated measuring segment.

6. A method as set forth in claim 1 wherein the sensor signals are respectively delivered by magnetic field sensors in the calibration run as well as in the measuring mode.

7. A method as set forth in claim 1 wherein the method further comprises the step of providing a sensor unit ( 7 ) including
a multiplexer ( 2 ), a single processing channel including an amplifier ( 3 ) and an analog/digital converter and a demultiplexer,
wherein the method further comprises the step of
feeding the respective signals of the plurality of sensor elements by means of said multiplexer alternately to said single processing channel, during the calibration run as well as in the me

Assignees

• Mehnert Walter
• Theil Thomas
• Avago Technologies General Ip

Inventors

• Mehnert Walter
• Theil Thomas

Classifications

• G01D5/145 Primary

  influenced by the relative movement between the Hall device and magnetic fields (see G01R33/06) · CPC title

• G01D5/14

  influencing the magnitude of a current or voltage · CPC title

• G01D18/002

  Automatic recalibration (G01D18/008 takes precedence) · CPC title

• G01D1/00

  Measuring arrangements giving results other than momentary value of variable, of general application (G01D3/00 takes precedence; in tariff metering apparatus G01D4/00; transducers not specially adapted for a specific variable G01D5/00) · CPC title

• G01D5/24452

  Physics · mapped topic