DE29612038U1 - Arrangement for linearizing the transfer function of a sensor - Google Patents

Description

Arrangement for linearizing the transfer function of a sensor
5

The invention relates to an arrangement according to the preamble of claim 1.

Analog inputs from measuring devices or input modules of controls or control systems are generally obtained as a coding of an analog/digital conversion that is proportional to an electrical quantity (e.g. in the form of a current, a voltage or a resistance). The values of these measured quantities are converted into a coding that is proportional to the physical quantity actually to be measured (e.g. in the form of a temperature, a pressure or a flow rate) before they are displayed or before further processing. The transfer function of the sensor, which describes the assignment of the electrical quantities to the physical quantities, is in some cases non-linear. In addition, there is the problem that the actual measurement is not specified by the transfer function itself, but by the inverse function. For example, for a large number of thermocouple types, the function U(T), i.e. the thermoelectric voltage as a function of the thermocouple temperature difference, is specified as an nth order polynomial with η = 4, ... 14.

In order to display a measured value corresponding to the value of the measured variable from a measured value processing unit, in particular a real-time measured value processing unit, one can proceed in such a way that the transfer function or the inverse transfer function is implemented in a linearization network and this is used in an approximation process to calculate the measured values. However, the computational effort is very high and floating point arithmetic is also required, which increases the processing time.

GR 96 G 4442 DE

The present invention is based on the object of creating an arrangement of the type mentioned at the outset which simplifies the determination of a measured value corresponding to the value of a measured quantity.
5

This object is achieved by the measures specified in the characterizing part of claim 1.

The advantage is that the measured value corresponding to the value of a measured quantity can be determined easily and quickly.

An advantageous embodiment of the invention results from the subclaim.

The invention is used in particular in analog input units.

The invention, its embodiments and advantages are explained in more detail with reference to the drawing, in which an embodiment of the invention is illustrated.

Show it:

Figure 1 shows an arrangement for linearizing the transfer function of a sensor and
Figure 2 shows the transfer function of a sensor in the form of a characteristic curve.

In Figure 1, 1 denotes a sensor to which an analogue measurement value in the form of a temperature T is fed. The sensor

1 generates from this temperature an auxiliary quantity in the form of a voltage U, which an A/D converter connected downstream of the sensor 1

2 into a &eegr; bit wide digital value UD. The A/D converter transmits this &eegr; bit wide digital value UD to a linearization network 3 in which the transfer function {Figure 2) of the sensor 1 is stored, which describes the assignment of the voltage U to the temperature T. The linearization network 3 also has means which are suitable for linearizing the transfer function of the sensor,

GR 96 G 4442 DE

and further means which display a measured value MW corresponding to the value of the measured quantity in a display and/or measured value processing unit 4.

To clarify the function and operation of the linearization network, reference is made to Figure 2, in which the transfer function of a sensor is shown in the form of a characteristic curve. The value range of the voltage U (auxiliary variable) from 0 to 160 μν is divided into 2 ^m equidistant areas, whereby in the present example m = 4 is selected, whereby 16 areas can be represented. The values T ₀ ... T ₁₅ of the measured variable T assigned to the range limits U ₀ ... U ₁₅ are stored as support points in memory cells of a memory of the linearization network 3 (not shown here).

It is assumed that the A/D converter 2 (Figure 1) supplies the linearization network 3 with an η = 8 bit wide digital value UD ₁ during measurement value acquisition. The corresponding support points are addressed with the m = 4 most significant bits Z, L·, 5_ and 4. of the digital value UD ₁ , whereby in the present example it is assumed that the digital value UD ₁ has the bit combination 1011 0000. With the most significant m (m = 4) bits Z, L·, ü and 4, which have the combination 1011, the linearization network 3 addresses the range limit U ₁₁ and thus the value T ₁₁ . The least significant bits 0., 1., 2. and 3. are each 0, whereby the value T ₁₁ represents the physical value which the linearization network 3 of the display and/or measured value processing unit 4 displays.

It is now assumed that a digital value UD _k has the bit combination 1100 1000. With the most significant m (m = 4) bits Z, £7 5th and 4th with the assignment 1100, the linearization network 3 addresses the range limit U ₁₂ and thus the value T ₁₂ , and with the least significant bits 3rd, 2, 1 and 0th with the assignment 1000, the linearization network 3 determines the distance to the determined value T ₁₂ .

GR 96 6 4442 EN

As described, the value range of the voltage U from 0 to 160 μν is divided into 16 equidistant areas, which means that each area has a voltage width of 10 μν. The least significant bit assignment 1000 therefore represents a voltage distance D ₁ of 5 μν according to a rule of three calculation, and the value of the measured quantity T ₁ to be determined is calculated as follows:

T -TT ₁ = -^ ^- * D.

Claims (3)

GR 96 G 4442 DE Protection claims 1. Arrangement - with a sensor (1) to which a measured variable (T) can be fed, from which the sensor (1) generates an auxiliary variable (U), which an A/D converter (2) converts into a &eegr; bit wide digital value (UD), - with a linearization network (3), - to which the A/D converter (2) supplies the digital value (UD), - which linearizes the transfer function of the sensor, which describes the assignment of the auxiliary variables (U) to the measured variables (T), and - which displays a measured value (MW) corresponding to the value of the measured quantity on a display and/or measured value processing unit (4), characterized in that - the value range of the auxiliary quantities (U) can be divided into 2 ^m equidistant areas, with m < n, - the values (T ₀ . . . T ₁₅ ) of the measured variables assigned to the range limits (U ₀ . . . U ₁₅ ) can be stored as support points in memory cells of a memory of the linearization network (3) which can be addressed with the m most significant bits of the η bit wide digital values (UD ₁ , UD _k ), - the linearization network (3) comprises means which - determine the corresponding support point from the m most significant bits of a received digital value (D ₁ , D _k ), - determine the distance (D _c ) to the determined support point from the remaining (n - m) bits of this digital value (D ₁ , D _k ) and - calculate the measured value (T ₁ ) from the measured value assigned to the determined support point and the determined distance. 2. Arrangement according to claim 1, characterized in that the value of a measured variable is determined according to the relationship GR 96 G 4442 DE _{γ =} y _k+ i - y _{k &khgr; + &eegr;} Jc+&igr; "~ Jt is calculated, where y _{± is} the i-th value of a measured quantity, y _k+1 is the value of the measured quantity at the (k+1) -th support point, y _{k is} the value of the measured quantity at the k-th support point, x _k+1 is the (k+1)-th support point, x _{k is} the k-th support point and X ₁ is the distance to the k-th support point. 3. Analog input unit with an arrangement according to claim or 2.