JP2973048B2 - Temperature sensor linearization processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a novel linearization processing method for a temperature sensor.

[0002]

2. Description of the Related Art Generally, a temperature control circuit for measuring an ambient temperature or performing control according to the ambient temperature includes:
A temperature sensor such as a thermistor or a platinum wire whose resistance value changes according to the temperature is used.

FIG. 7 is a characteristic curve showing a change in resistance value with respect to the ambient temperature for a thermistor having a negative temperature coefficient. Usually, it is necessary to obtain the temperature from the resistance value of the thermistor. By using the bridge circuit or the like shown in FIG. 8, linearization is performed so that the output voltage proportional to the temperature shown in FIG. 9 is obtained, and the obtained voltage is used to indicate a meter or to obtain the output voltage. The converted voltage was converted to digital data for temperature display and control. However, since a large number of resistors R are used in the bridge circuit shown in FIG. 8, it has been difficult to suppress a linearization error of the circuit caused by a variation in the resistance value R or the like.

FIG. 10 is a characteristic curve showing a change in resistance value of a platinum temperature sensor with respect to an ambient temperature.
Although it has a positive temperature coefficient on a substantially straight line, in this case also, it is usually necessary to obtain the temperature from the resistance value of the platinum temperature sensor.
The linearization is performed by the bridge amplifier circuit using the circuit so that an output voltage proportional to the temperature can be obtained as shown in FIG. 12, but this bridge amplifier circuit configuration is similar to the circuit shown in FIG. Since a large number of resistors R are used, it is difficult to suppress a circuit error caused by a variation in resistance value and the like, which makes circuit design difficult.

[0005]

SUMMARY OF THE INVENTION The present invention has been proposed in view of the above circumstances, and has been proposed to reduce the temperature measurement error by a simple method and to reduce the load of arithmetic processing. It is intended to provide a way.

[0006]

According to the first aspect of the present invention, which is proposed to achieve the above object, a resistance characteristic of a temperature sensor with respect to temperature is approximated by a hyperbola for each predetermined section,
A linearization table storing resistance value ranges and hyperbolic approximations corresponding to the classification is prepared in advance, and when sampling the resistance value of the temperature sensor, the resistance value is referred to by referring to the linearization table. The corresponding hyperbolic approximation formula is read, and the sampled resistance value is substituted for the read hyperbolic approximation formula to calculate the temperature value. According to a second aspect of the present invention, a thermistor having a negative temperature coefficient is used as the temperature sensor. In the present invention, a platinum wire having a positive temperature characteristic is used as the temperature sensor.

[0007]

According to the first aspect of the present invention, the resistance characteristic of the temperature sensor with respect to the temperature is approximated by a hyperbola for each predetermined section, and the resistance value range corresponding to the section is stored in correspondence with the hyperbolic approximation formula. A prepared linearization table is prepared in advance. Then, when the resistance value of the temperature sensor is sampled, a hyperbolic approximation formula corresponding to the sampled resistance value is read out with reference to the linearization table, and the sampled resistance value is substituted into the read-out hyperbolic approximation formula to obtain the temperature. Calculate the value. Therefore, if a linearization table is prepared in which the resistance characteristic of the temperature sensor with respect to the temperature is accurately approximated by a hyperbola, errors can be reduced and the linearization processing in which the arithmetic processing is simplified can be performed.

According to the second aspect of the present invention, a thermistor having a negative temperature characteristic is used. Therefore, when it is desired to make the output voltage characteristic with respect to the temperature in an inversely proportional relation, or due to the nature of the thermistor, measurement of a low temperature is performed. Suitable for such as.

Further, according to the present invention as set forth in claim 3,
Since a platinum wire having a positive temperature characteristic is used, it is suitable for a case where the output voltage characteristic with respect to the temperature is made to be in a direct proportional relation, or for the measurement of a high temperature due to the nature of the platinum wire.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1A is a flowchart showing a linearization processing method according to the present invention when a thermistor is used as a temperature sensor. The processing according to the present invention will be described with reference to the mathematical formula in FIG. The method will be described.
Note that a method for creating a linearize table, which is the basis of the present invention, will be described later. The resistance value R1 of the thermistor TH is sampled. As a sampling method, as shown in FIG. 1B, a power supply voltage V1 is applied to a thermistor TH and a resistor r connected in series, and a terminal voltage V2 of the thermistor TH at that time is sampled. , V2 / V1 = R1 /
From the proportional relationship of (r + R1), R1 is obtained as in the following equation. R1 = (V2 / (V1−V2)) × r (1) Referring to the linearization table, it is expressed by the following equation corresponding to the resistance R1 of the thermistor TH calculated by the equation (1). Read the hyperbolic equation T = (a1 / (R-b1)) + c1 (2) The temperature Ts is calculated by substituting the resistance R1 of the thermistor TH into the read hyperbolic equation (2). Ts = (a1 / (R1-b1)) + c1 (3) (see steps (a) and (100) in FIG. 1).

As described above, according to the method of the present invention, only by preparing a linearization table as described later in advance, as shown in equation (3), one addition, one subtraction, and one division Since the temperature value Ts can be calculated by a simple equation, the resistance value characteristic of the thermistor with respect to the temperature can be calculated as 2
The burden of arithmetic processing is reduced compared to complicated processing that approximates a next curve, etc. In addition, since a large number of resistors are not used unlike the conventional bridge circuit configuration, errors caused by variations in resistance are reduced. Can be done.

Next, a method of preparing a linearize table will be described with reference to the resistance value characteristics with respect to temperature of the thermistor shown in FIG. To approximate the resistance characteristics in the temperature range 0-10 ° C,
The resistance values at 0 ° C, 5 ° C and 10 ° C are determined. As a result, 0 ° C → 65KΩ, 5 ° C → 52KΩ,
10 ° C → 42 KΩ is obtained. The obtained value is converted to the hyperbolic basic formula T = (a / (R−b)) +
Substitute for c. That is, 0 [゜ C] = (a / (65 [KΩ] -b)) + c (4) 5 [゜ C] = (a / (52 [KΩ] -b)) + c ... · (5) 10 [゜ C] = (a / (42 [KΩ]-b)) + c · · · · · · · · · · · · Formulas (4) to (6) are simultaneously calculated to obtain a, b, and c. Thus, an approximated hyperbolic equation is obtained. When a, b, and c are actually obtained from Expressions (4) to (6), a = 33.
22221, b = -34666, c = -33.3, and the hyperbolic approximate expression L1 in the temperature range of 0 to 10 ° C. is as follows. Ts = (3322221 / (R1 + 34666)) + 33.3 (7) The obtained hyperbolic approximation equation (7) is converted into a resistance value range corresponding to a temperature range of 0 to 10 ° C., that is, 65 [KΩ] to 42
It is stored in the linearization table together with the data of [KΩ]. The above-mentioned processes (1) to (10-20 ° C, 20-3)
The hyperbolic approximation formulas L2 to L5 are sequentially obtained by repeatedly performing each of the temperature ranges of 0 ° C, 30 to 40 ° C, and 40 to 50 ° C. The corresponding expression is stored in the linearization table.

FIG. 3 is a graph in which the error between the calculated temperature value obtained by the above-described method of the present invention and the actual temperature value is plotted . For each temperature range approximated by a hyperbola, +0. The temperature can be kept within an error of 03 [゜ C] to -0.03 [゜ C ], and high-precision temperature measurement can be performed.

FIG. 4A is a flowchart showing the linearization processing method of the present invention when a platinum temperature sensor is used, and FIG. FIG. 1A and FIG. 1B are only different from those in FIGS. 1A and 1B in that a resistance value is sampled.
The description of the other same processes is omitted. The resistance value R1 of the platinum temperature sensor PT is determined by applying a constant current I from the constant current circuit C to the platinum temperature sensor PT, measuring the terminal voltage V1, and calculating the resistance value R1 from R1 = V1 / I. Have been.

FIG. 5 is an explanatory diagram in the case where the resistance characteristic of the platinum temperature sensor with respect to the temperature is approximated by a hyperbola. The procedure for obtaining the hyperbolic approximation is the same as that in FIG.

FIG. 6 shows a graph obtained by measuring and plotting the error between the temperature calculated by the platinum temperature sensor obtained by the method of the present invention and the actual temperature, and shows each temperature range approximated by a hyperbola. +0.02 [゜ C] to-
The temperature can be kept within an error of 0.02 [゜ C] , and high-precision temperature measurement can be performed.

In the above description, the case where the hyperbolic approximation formula is obtained by dividing the temperature range every 10 degrees is described as an example. However, the temperature range is further finely divided and more accurate hyperbolic approximation is performed. It is also possible to have it done.

[0018]

As will be understood from the above description, according to the present invention, a linearization table in which the characteristics of the temperature sensor are approximated to a hyperbola is prepared in advance. By simply setting, the temperature can be accurately calculated from the resistance value of the temperature sensor by a simple calculation process. For this reason, it is possible to reduce the load of the arithmetic processing as compared with the case where the characteristics of the temperature sensor are approximated to a quadratic curve or the like, and it is possible to perform high-precision measurement in which measurement errors caused by errors such as resistance are reduced.

[Brief description of the drawings]

FIGS. 1A and 1B are explanatory diagrams of a processing method of the present invention when a thermistor is used as a temperature sensor.

FIG. 2 is an explanatory diagram of a method of creating a linearization table when a thermistor is used as a temperature sensor.

FIG. 3 is an explanatory diagram of a measurement error when a temperature is measured by the processing method of the present invention shown in FIG. 1;

FIGS. 4A and 4B are explanatory diagrams of a processing method of the present invention when a platinum temperature sensor is used.

FIG. 5 is an explanatory diagram of a method for creating a linearize table when a platinum temperature sensor is used.

6 is an explanatory diagram of a measurement error when performing temperature measurement by the processing method of the present invention shown in FIG. 4;

FIG. 7 is a graph showing a resistance value characteristic with respect to a temperature of a thermistor.

FIG. 8 is a conventional linearize circuit applied to a thermistor.

9 is an explanatory diagram of an output signal obtained by the linearize circuit shown in FIG.

FIG. 10 is a graph showing a resistance value characteristic of a platinum temperature sensor with respect to a temperature.

FIG. 11 is a conventional linearize circuit applied to a platinum temperature sensor.

12 is an explanatory diagram of an output signal obtained by the linearize circuit shown in FIG.

[Explanation of symbols]

 TH: Thermistor PT: Platinum temperature sensor

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G01K 7/00 G01D 3/02

Claims (3)

(57) [Claims] 1. A linearization table in which resistance characteristics of a temperature sensor with respect to temperature are approximated by a hyperbola for each predetermined section, and a resistance value range corresponding to the section and a hyperbolic approximation equation are stored in advance. When sampling the resistance value of the temperature sensor, the hyperbolic approximation formula corresponding to the resistance value is read with reference to the linearization table, and the sampled resistance value is substituted into the read hyperbolic approximation formula to calculate the temperature value. And a linearization processing method for the temperature sensor. 2. A method according to claim 1, wherein said temperature sensor comprises a thermistor having a negative temperature coefficient. 3. The method according to claim 1, wherein said temperature sensor comprises a platinum wire having a positive temperature coefficient.