Arrangement and process for interpolating a measured signal

Details Arrangement and process for interpolating a measured signal Arrangement and process for interpolating a measured signal

2002-12-06

2004-05-25

Williams, Howard L. (Department: 2819)

Coded data generation or conversion

Analog to or from digital conversion

Analog to digital conversion

C341S111000

Reexamination Certificate

active

06741199

ABSTRACT:

BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The present invention relates to an arrangement and process for interpolating a measured signal.
Sensor systems with incremental and absolute measurement and increasingly, for example for angular position and position measurements, the combination of both measurement techniques in high resolution absolute measuring systems are used for industrial purposes. There is a continual requirement for recording absolute position and relative displacement in measuring systems but it is often limited to purely relative measurements since for example the incremental sensor systems suited to this purpose are more cost-effective to produce.
We explain below the basic requirements for high resolution measured signals based on the digitalization of angular position and position measuring systems, whereby our observations may equally be attributed and applied to other physically/chemically derived measured values.
In order to use an incremental measuring system to record absolute position as well, other auxiliary units are required, which determine on the one hand the absolute starting position and permanently calculate on the other hand the incremental signals relative to it sequentially and directionally. Besides the ON/OFF effect of the supply voltage, interference to the mains or faults in transfer of the incremental signals also cause a loss of the absolute signal value in the need to establish a new reference each time. For machines in continuous operation in environments which are prone to interference in particular these measuring systems are unreliable and problematic in an otherwise uninterrupted production process.
The conventional absolute measuring system is again very costly in the implementation of its sensor system and reaches its limits, particularly where high resolutions of the angular and linear distances traveled also in the high speed range are required.
Combined measuring systems have therefore increasingly become established and have absolute angular position/position measuring sensors with resolutions of for example ≧8 bits, i.e. angular distances which are smaller than 1.5° and linear distances of ≦50 &mgr;m.
Useful measured signals totally converted to absolute values in the angular position/position measuring sensor—as described in more detail in the application for Swiss patent CH210599—require in particular a careful method of recording and evaluating the measured signals. Between the absolute values which are determined separately in these systems there are incremental values lying between two absolute values which are used to create the total absolute value. The analog-digital converters with comparator technologies used for interpolating the incremental values have the disadvantage that the cost for resolutions of even 5 to 6 bits are too high and the required hardware integration approaches its limits.
For every step increase in resolution, the cost of hardware integration increases by a factor of 2. This technology is shown in
FIG. 1
of the drawing. The measured signal U
1
can change e.g. linearly from 0 up to U
1max
. U
1
is applied to a voltage divider Sp, whose pickoffs are all coupled individually to an operational amplifier COMP, to which the reference voltage U
2
is also fed. Each of the operational amplifiers COMP switch, whenever the pickup voltage fed to them has reached the preset threshold. The switching of the operational amplifiers determines each particular intermediate value reached, which is then added to the neighboring absolute value.
The above mentioned patent specification CH210599 points to the expected increase in processing speed of semiconductor technology as a practicable solution to overcoming the high cost of higher resolutions and suggests carrying out until that time additional sine/cosine signals and undertaking higher resolution AD conversion externally.
SUMMARY OF THE INVENTION
The object of the present invention is to create an interpolation circuitry as well as a process for interpolating analog measured variables which operates with higher speed and accuracy and overcomes the above mentioned disadvantages.
This object is achieved by an arrangement for interpolating an analog measured signal which is dependant on the measured variable, and which is fed to a voltage divider (Sp) whose pickoff signals are fed to comparators, from whose output signals the interpolated measured variable is determined, wherein the pickoff signals are fed by means of switches (S
1
to S
8
) to comparators (Comp, Comp
−1
) whose number is smaller than that of the pickoff signals and the switches (S
1
to S
8
) connect the pickoff signals consistently to comparators (Comp, Comp
−1
).
It is generally understood that by comparator is meant a suitable threshold switch which operates sufficiently fast and with sufficiently high resolution.
The interpolator or interpolation according to the invention makes particular use of the continuity of functions for measured signal processing used for determining measured variables, e.g. with trigonometric functions for SIN and COS signals. Below we explain the beneficial design of the interpolator or interpolation method using the example of SIN/COS signal processing of position-dependent variables, whereby the arrangement as well as its implementation is basically suitable for and is of benefit for use in other analog and especially continuous analog measured signal flows.
A sensor system must record measured values as continuously as possible if it is to be used also in dynamic applications. This ensures the fastest possible access to measured values at any time since they are already processed when they are called up and may be outputted in “real time”. Special measures must be taken with such “real time” measuring systems whose clock-pulse rate for logical signal processing is also used for interpolating analog values. Apart from the increasing cost of for example hardware logic to interpolate higher resolutions, it is particularly important to take into account the ever smaller signal amplitudes for each step in resolution critical for accuracy and sensitivity of the recording process in addition to time-critical switching procedures.
The object of the invention is also achieved by a method for interpolating an analog measured signal which is dependant on the measured variable, and which is fed to a voltage divider (Sp) whose pickoff signals are fed to comparators, from whose output signals the interpolated measured variable is determined, wherein the pickoff signals are fed by means of switches (S
1
to S
8
) to comparators (Comp, Comp
−1
) whose number is smaller than that of the pickoff signals and the switches (S
1
to S
8
) connect the pickoff signals consistently to comparators (Comp, Comp
−1
).
The method of interpolation described in the invention solves the above-mentioned object by selecting a smaller number of comparators to use for comparison than the number of reference points given by the resolution and by using switches to feed these consistently to at least one comparator for signal evaluation.
This beneficial method of interpolation has limits in the resolution of the analog values given by the accuracy of the resistor dividers to be formed and the offset values of the comparators. Since the resistance values are in the thousandths range of the accuracies which can be achieved by semiconductors and the offset values of the comparators are also in the thousandths range of the useful signal, resolutions of approx. 8-bit can still be achieved at the first stage of interpolation without a great deal of extra expenditure. The advantage of the reduced number of comparators can find an additional use, however, in that an on chip offset adjustment is carried out which then allows the range of resolution to be expanded up to 10-bit with correspondingly more expensive resistors—e.g. parallel switching of the resistors allows a corresponding increase in accuracy.
In many applications, e.g. servo drives for position con

Affiliated with

Also associated with

Rating

Arrangement and process for interpolating a measured signal does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Arrangement and process for interpolating a measured signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Arrangement and process for interpolating a measured signal will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3203359