JPS6350648B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an improved method for measuring the temperature of a fluid.

When using a temperature sensor to measure fluids such as gases and liquids whose temperature changes over time, the temperature sensor has a heat capacity, so the temperature of the temperature sensor will be delayed by a certain amount of time relative to the temperature change of the fluid being measured. The response will be as follows. For this reason, when measuring fluid temperatures that fluctuate rapidly, conventionally,
Measurements were performed using a circuit that compensated for response delays using a fixed time constant for the output electrical signal from a single temperature sensor. However, even with such conventional methods, it is not possible to measure the temperature of a fluid in an accurate temperature and time relationship. That is, when the velocity, temperature, density, etc. of the fluid to be measured changes over time, the time constant of the response of the temperature sensor also changes, but in the conventional technology, measurement errors occur because response compensation is performed using a fixed time constant. Furthermore, in order to set the time constant, it is necessary to measure the time-average velocity, temperature, density, etc. of the fluid to be measured in advance and then calculate the time constant, or to actually measure the time constant.

The present inventor has completed the present invention as a result of intensive research to solve the above-mentioned problems encountered in fluid temperature measurement.

That is, according to the present invention, two temperature sensors w ₁ and w ₂ of the same type with different dimensions and similar shapes are inserted into the part where the temperature of the fluid is measured, and two electrical signals corresponding to the temperature of each temperature sensor are generated. A fluid temperature measuring method is provided, which is characterized in that the fluid temperature T _f is calculated by obtaining T _w1 and T _w2 and using these electric signals to perform the calculation shown by the following formula.

T _f = dT _W2 /dt・T _w1 − (d _W1 /d _W2 ) ⁿ・dT _W1 /dt・T _w2 /d
T _W2 /dt−. (d _W1 /d _W2 ) ⁿ・dt _W1 /dt(1) (Each symbol in the formula means the following.

T _f ...Fluid temperature T _w1 ...Electrical signal corresponding to the temperature of temperature sensor _w1 T _w2 ...Electrical signal corresponding to the temperature of temperature sensor _w2 d _Tw1 /dt...Differential value of T _w1 with respect to time d _Tw2 /dt... Differential value of T _w2 with respect to time d _w1 /d _w2 ... Size ratio of temperature sensors w ₁ and w ₂ n... Number from 1.5 to 1.9) As is clear from the above equation (1), In the present invention, unlike the conventional method, when measuring the fluid temperature T _f , the calculation formula (1) uses two temperature sensors, so that the fluid temperature, temperature, density, etc. No factors are included. Therefore,
According to the present invention, by simply measuring the dimensional ratio of the two temperature sensors in advance, it is possible to accurately measure the fluid temperature in which the response delay at each moment is reliably compensated for.

Next, the measurement principle of the present invention will be explained with reference to the drawings.

Figure 1 is an explanatory diagram showing heat transfer due to contact between a temperature sensor and a fluid to be measured, where f is the fluid to be measured, T _f is the fluid temperature, w is the temperature sensor, d _w is the diameter of the temperature sensor, and l _w is the length of the temperature sensor,
T _w indicates the temperature of the temperature sensor.

Now, if the temperature sensor is made cylindrical as shown in Figure 1, and heat radiation and conduction are ignored, the change in the amount of heat held by the temperature sensor will be the same as the temperature sensor w and the fluid f.
Since it is equal to the amount of heat transferred between Note that the shape of the temperature sensor does not have to be cylindrical as shown in the drawings, and may have other shapes, but here, it is assumed to be cylindrical for ease of explanation. Also, the temperature of the fluid is 1000
It has been theoretically and experimentally confirmed that thermal radiation and conduction can be ignored as long as the temperature is not much higher than ℃ and the diameter is small compared to the length, as is the case with ordinary temperature sensors. ing.

c _w ρ _w l _w π/4d _w ² dT _w /dt = πd _w l _w α(T _f −T _w ) (2) Each symbol in the formula means the following.

c _w …Specific heat of temperature sensor ρ _w …Density of temperature sensor l _w …Length of temperature sensor d _w …Diameter of temperature sensor T _w …Temperature of temperature sensor T _f …Temperature of fluid t …Time α …Fluid and temperature Heat transfer coefficient between the sensor and the sensor Next, by transforming equation (2), it becomes T _f = T _w + c _w ρ _w d _w /4α・dT _w /d _t (3), and the heat transfer coefficient α is α = Nuλ /d _w (Nu: Nutsselt number, λ: thermal conductivity of fluid), T _f =T _w +c _w ρ _w d _w ² /4Nuλ・dT _w /dt = T _w +τdT _w /dt . τ in equation (4) defines the response delay of the temperature sensor to the temperature change of the fluid to be measured, and is called a time constant.

Nu is generally expressed as Nu=A+BRe ^p Pr ^q =A+B(d _w u/ν) ^p (ν/a) ^q (5) Nu=CRe ^p Pr ^q =C(d _w ^u /ν) ^p (ν/a ) ^q (6) etc. Here, A, B, C, p, q
are constants proposed by various researchers, Re...Reynolds number Pr...Prandtl number u...Velocity of fluid ν...Kinematic viscosity coefficient a of fluid...Temperature conductivity of fluid . λ, ν, a, etc. are physical constants determined by the type of fluid and temperature, and c _w also changes slightly depending on the temperature. Therefore, if the velocity or temperature of the fluid to be measured changes over time, the time constant τ also changes.

In the conventional technology, the time constant τ is set to a constant value and the fluid temperature that changes over time is measured by calculation using equation (4), and the setting of τ is determined experimentally by calculation. There are two ways to do it.

When determining by calculation, measure the average velocity and average temperature of the fluid in advance, and calculate λ, ν, a
The physical property constants are obtained from the thermophysical property data, and calculated from the sensor's c _w , ρ _w , and d _w as τ = c _w ρ _w d _w /4Nuλ.

When determining this experimentally, the sensor is inserted into the fluid to be measured, and the temperature of the sensor is raised above the temperature of the fluid by a method such as resistive heating by passing an electric current through the sensor. The time constant τ is determined by measuring the subsequent transient response of the sensor temperature. However, since the fluid temperature fluctuates over time, this operation must be repeated several times to obtain the average τ.

Therefore, the conventional technology not only requires troublesome preparations for setting τ, but also assumes a constant τ even though τ actually changes over time. (In gas-liquid two-phase flow and combustion gas, physical property values also change depending on the type of fluid.) The error is significant when the change is large.

In the present invention, homogeneous temperature sensors with different diameters (diameters are d _w1 and d _w2 , respectively) arranged at small intervals are used. Then, similarly to equation (4), holds, but c _w ρ _w /4λ can be eliminated from these two equations: T _f = dT _W2 / dtT _w1 − (d _W1 / d _W2 ) ² (N _u2 /N _u1 )・dT _W1 /
dtT _W2 /dT _W2 /dt−(d _W1 /d _W2 ) ² (N _u2 /N _u1 )・dT _W1 /
dt (8) is obtained. When we examine various experimental formulas expressed by equations (5) and (6) for Nu, we find that in all of them, the relationship (Nu ₂ /Nu ₁ )≒(d _w2 /d _w1 ) ^m (9) is established with extremely small errors. is found. Therefore, T _f = dT _W2 /dtT _w1 − (d _W1 /d _W2 ) ⁿ・dT _W1 /dtT _w2 /dT _W2
/dt−(d _W1 /d _W2 ) ⁿ・dT _W1 /dt(1). Here, n = 2 - m, but as can be seen from equation (9), (d _w2 /d _w1 ) ^m is determined by the Nutsselt number ratio Nu ₂ /Nu ₁ shown in equation (5) or (6). In general, a number in the range of 1.5 to 1.9 is appropriately adopted as n depending on the measurement conditions, but in normal cases, a sufficiently accurate measurement value can be obtained by adopting a number of 1.7.

In the present invention, λ, ν, which change over time,
Since values such as a are common to the two sensors,
In other words, the temporal change in τ is the difference in the change in T _w1 and T _w2 (i.e., when τ is small, the difference between T _w1 and T _w2 is small, and when τ is large, the difference between T _w1 and T _w2 is large). This means that it is accurately taken into account.
In addition, the only constant used when performing calculations is (d _w1 /d _w2 ), which does not depend on the type of fluid, flow rate, or temperature, and there is no need to prepare for each measurement in the prior art. In the present invention, the temperature sensor may be anything that can convert temperature into an electrical signal, such as a thermocouple or a resistance thermometer.

To carry out the method of the invention, two homogeneous temperature sensors of different dimensions and similar shapes are placed in the measuring portion of the fluid to be measured (for example, a liquid, a gas, or a mixture of both); The temperature of the sensor is expressed as an electric signal T _w1 which is proportional to each temperature.
Convert to T _w2 . In this case, the two temperature sensors are placed as close as possible (usually less than 1 mm) so that they contact the same temperature portion of the fluid.
Next, using the electric signal thus obtained, the above equation (1) is calculated to obtain the fluid temperature T _f .
The shape of the temperature sensor does not have to be exactly similar, but may be roughly similar, such as a cylinder or a prism. Further, the size ratio may be a typical value for the shape of the sensor; for example, in the case of a thermocouple, the wire diameter ratio is used.

In the present invention, the calculation of equation (1) can be performed by an analog circuit, or can be performed by a computer after A/D conversion of the output signal from the temperature sensor.

Next, the present invention will be explained in more detail with reference to Examples.

EXAMPLE Two temperature sensors were placed in the flame to measure the temperature of the hot gas generated when propane burns.

Temperature sensor (w ₁ )...Tanaka Kikinzoku Kogyo Co., Ltd. thermocouple (JIS R type) diameter 50 μm, length 30 mm Temperature sensor (w ₂ )...Tanaka Kikinzoku Kogyo Co. thermocouple (JIS R type) diameter 75 μm, length 30 mm Size ratio (d _w1 / d _w2 ) = 50 / 75 = 0.667 Interval between thermocouples (w ₁ , w ₂ )...0.5 mm Using a computer, calculate the electrical signals obtained from the two thermocouples. Perform arithmetic processing according to the above formula (1) (in this case, 1.7 is adopted as n in formula (1)),
The results obtained are shown graphically in FIG.

For comparison, measured temperatures obtained from thermocouples w ₁ and w ₂ without compensation for response delay are shown as graphs in FIGS. 3 and 4, respectively.

As can be seen by comparing FIG. 2 with FIGS. 3 and 4, it is clear that according to the present invention, instantaneous temperature changes that could not be detected due to response delay can be detected with high sensitivity.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram for explaining heat transfer due to contact between a fluid and a temperature sensor. Figure 2 shows
3 is a graph showing temperature measurement results obtained by the present invention. Figures 3 and 4 are graphs showing the temperature measurement results directly obtained from the thermocouple without compensation for response delay, and Figure 3 is the graph showing the temperature measurement results obtained directly from the thermocouple.
w ₁ (d _w1 = 50 μm) and Figure 4 shows the temperature sensor w ₂
(d _w2 =75 μm). f...Fluid, T _f ...Fluid temperature, w...Temperature sensor,
T _w ...Temperature sensor temperature, d _w ...Temperature sensor diameter, l _w ...Temperature sensor length.

Claims (1)

[Claims] 1. Two temperature sensors w ₁ and w ₂ of the same type with different dimensions and similar shapes are inserted into the part where the temperature of the fluid is measured, and two electrical signals T corresponding to the temperature of each temperature sensor are generated. A method for measuring the temperature of a fluid, which comprises obtaining _w1 and T _w2 and calculating the fluid temperature T _f by using these electric signals and performing the calculation shown by the following formula. T _f = dT _W2 /dt・T _w1 − (d _W1 /d _w2 ) ⁿ・dT _W1 /dt・T _w2 /d
T _W2 /dt−(d _W1 /d _W2 ) ⁿ・dT _W1 /dt (Each symbol in the formula means the following. T _f ...Fluid temperature T _w1 ...Corresponds to the temperature of temperature sensor _w1 Electric signal T _w2 ...Electrical signal corresponding to the temperature of temperature sensor w ₂ dT _w1 /dt ... Differential value of T _w1 with respect to time dT _w2 /dt ... Differential value of T _w2 with respect to time d _w1 /d _w2 ... ...Dimension ratio n of temperature sensors w ₁ and w ₂ (number from 1.5 to 1.9).