JPS5925964B2 - Temperature measurement device using resistance thermometer - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is directed to an apparatus for measuring and monitoring the temperature of marine engines and the like.

Conventionally, temperature measurements using resistance temperature sensors such as platinum,
For each measurement point, construct a bridge circuit with a temperature sensing resistor on one side, amplify its output, and use the following (C) approximation formula to feedback the output of the bridge amplifier to the gain using the linear method or the bridge circuit. The measured temperature values were obtained by the linear rice method.

FIG. 1 is a connection diagram showing a typical example of a feedback type bridge linear rice circuit. The resistance value of the temperature measuring resistor when the temperature is T is expressed by the first equation. to=to0(1+αT
+βT2) ... (1) However, ro is the initial resistance value of the resistance temperature detector, α and β are α■3.98× at 0 to 600℃
10−, β=−6.13×10−7. However, these cases have the following drawbacks.

(1) For each measurement point, a bridge resistor element that requires relatively high precision and high stability, a bridge amplifier circuit that amplifies the bridge output, and a linear rice that linearizes the correspondence between the temperature of the measurement point and the output value. Requires circuit and hardware.
(2) Initial adjustment of the bridge amplifier circuit and linear rice circuit and correction of changes over time are required.

Particularly when the number of measurement points is large, the hardware cost and adjustment time are enormous, and the reliability and stability of the measurement are also reduced. In order to eliminate these drawbacks, the present invention simplifies the hardware required per measurement point to one reference resistance element, and makes various necessary corrections based on line resistance, ambient temperature, reference voltage value, etc. By separately obtaining correction data, arithmetic processing is performed to obtain a temperature measurement value with a predetermined accuracy.

Hereinafter, the hardware configuration of the embodiment of the present invention, the introduction of the temperature resistance relational expression of the platinum resistance temperature sensor, the analysis of each error term, the correction calculation process, and the accuracy will be described in detail.

FIG. 2 shows a schematic system diagram of the entire temperature measurement and monitoring device according to one embodiment of the present invention.

Signals from multiple resistance temperature detectors (d front end section 2
A multiplexer (M
PX) 5 and converted into a digital value by the AD converter 6. The arithmetic processing circuit 7 calculates the measured temperature value by linear rice processing and various correction processes, and outputs the result to the printer 8. FIG. 3 shows the front end section of D per measurement point.

When a current is passed from the reference power supply V to the temperature sensing resistor r via the reference resistor R, the total voltage drop Vr including the line resistance a-a'', cmc'' and the voltage drop due only to the line resistance cmc' vδ
Two signals are measured at each measurement point. FIG. 4 is a simple resistance division circuit diagram using resistance elements Rl and R2 with matched temperature characteristics for detecting a variation ΔV in the reference power supply voltage V.

The ambient temperature value t for correcting the measurement value error caused by the temperature coefficient of the reference resistance is obtained using a current output type clogging sensor C or the like having the required resolution and linearity only in the room temperature range.

FIG. 5 is a circuit diagram showing an example thereof. When deriving the temperature from the resistance value, equation (1) still has a large error in relation to the actual characteristics, especially on the high temperature side, which has an approximation error of 0.5°C or more, so the value to be corrected is about the same. The accuracy is insufficient as a temperature-resistance relational expression that serves as the basis for finely correcting each error factor.

Therefore, in order to improve the approximation accuracy in this high temperature region, we introduce the following relational expression that performs approximation using a hyperbola including trigonometric relationships. TllO ~ 600℃ A=2444.060ΩB-137
63660Ω·°C C=5871.705°C Z=0.26rc As a result, the difference from the actual characteristics is 0.05°C or less at 0 to 400°C and 0.08°C or less at 400 to 600°C.

When determining the measured temperature T based on equations (2-1) and (2-2), the actual variable r is determined by the reference voltage V, reference resistance R, and acquired data Vr and vδ.

Therefore, equation (2-2) is converted into variables and becomes the third equation. T
A=f(r)=G(V,R,vδ,Vr)...
...(3) Here, as error factors, reference power supply fluctuation DV,
Considering the initial deviation DRi of the reference resistance, the variation DRt of the reference resistance due to the ambient temperature, and the potential Dvδ due to the line resistance.

The actual correction calculation is performed in the following steps.

First, as Dvδ,d multiplied by the AD conversion factor? =2000>(DvδμV22000×DV are defined respectively.

Next, regarding the ambient temperature that determines DRt, the deviation of the t sensor data from 25°C, the temperature coefficient of resistance, and its initial deviation are determined to be 32X△T, lO'5×β, respectively, in order to equalize the resolution.
, 101×ΔR. As a result, the correction formula (4-2) has the following form in which the contribution of each term, which is advantageous in actual digital calculations, is balanced. DT-kδ・? +Kv
- Base KRi・Sphere+KRt・β・Δt...Each coefficient is
(5-1) However, pull R'
A=2.3041475×104N-(A+RO)/A
= 1.40915526 x = 2000

(1) The coefficients kδ, Kv, kRi, kRt are not calculated each time, but are 0.
In order to obtain an accuracy of about 3°C, a coefficient table is set in memory for every 128 steps regarding X.

(2) Set the table of β and ΔR from the resistance element to be used in the memory. (3) Determine DT from the above numerical table by multiplication and summation of 16 BIT.

(4) For TO, set a table for each step regarding X in memory.

(5) T.O. l! : Determine the final measured feminine power value T by the sum with DT.

(6) As for the final accuracy, an accuracy of about 0.4° C. is obtained as the sum of the approximation accuracy of the degree-resistance relational expression of equations (2-1) and (2-2) and the correction calculation accuracy.

Measure as explained above. Temperature measurement using a flat resistor is extremely simple per measurement point without using complicated bridge circuits, linear rice circuits, etc. - All it takes is a low-cost microcomputer and a small amount of memory for batch correction processing. This makes it possible to achieve accurate measurement at a sufficiently practical sampling rate, which simplifies the arm measurement device. It can be made smaller and has a particularly significant effect on multi-point measurement. Furthermore, since no adjustment is required, the manufacturing process is simplified and long-term measurement reliability is achieved.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a connection diagram of the front end of a conventional temperature measuring device.

Claims (1)

[Claims] 1. A plurality of resistance temperature detectors, a reference resistor connected in series to each of the resistance temperature detectors, a reference voltage source and cable that apply a voltage to the reference resistance and the resistance temperature detector, and the reference voltage source. a V sensor that detects voltage fluctuations, a t sensor that measures ambient temperature, and a voltage drop (v
r) and the voltage drop (vδ) of the cable by sequentially switching and detecting the voltage drop (vδ) and outputting it together with the t sensor output and the V sensor output, an AD converter that AD converts each output of the multiplexer, and an AD converter that converts each output of the multiplexer into an AD converter; A temperature measuring device comprising: an arithmetic device that performs hyperbolic approximation calculations including trigonometric functions using a memory that stores correction coefficients.