JPS5821214B2 - temperature measuring device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an element (hereinafter referred to as a conversion resistor) that converts temperature into an electrical resistance value in response to temperature.

The present invention relates to a technology for measuring temperature using a resistance/pulse width conversion circuit that converts the resistance value of ( ) into a pulse width.

When measuring temperature remotely, a conventional method is to convert the temperature to be measured into an electrical quantity using a primary converter, convert the temperature to be measured into a pulse width using this electrical quantity, and transmit this pulse signal. It is being

In this case, the circuit that converts the electrical quantity into a pulse width is required to operate sufficiently stable even if external conditions change, and the Good linearity is required between the amount of electricity and the pulse width.

FIG. 1 shows a conventional example of a temperature converter using a resistance/pulse width conversion circuit used in such cases.

In FIG. 1, R1 is a conversion resistor whose resistance value changes depending on the temperature, and R2, R3, and R4 are fixed resistors, respectively, and these resistors R1, R2, R3, and R4 form a bridge circuit.

Also, 1 is a DC power supply with a voltage of E volts, 2 is a DC amplifier,
3 is a voltage-controlled pulse width converter whose pulse width changes depending on the input voltage, and 4 is a pulse generator whose repetition rate is constant.

For the conversion resistor R11, a temperature-sensitive resistance wire such as an iron-nickel alloy wire or a platinum resistance wire, a thermistor, or the like is used.

When the conversion resistor R1 changes in accordance with the temperature to be measured, the output voltage Vj of the bridge circuit changes accordingly, and this output voltage Vi is amplified by the DC amplifier 2 and input to the pulse width modulator 3, and is then input to the pulse generator. The pulse width of the pulse supplied from 4 is changed to obtain a pulse having a time width ts corresponding to the temperature to be measured.

The measurement circuit shown in FIG. 1 measures the voltage E of the linear power supply 1, the amplification degree and linearity of the DC amplifier 2, the operation of the pulse width modulator 3 and the pulse generator 4, the temperature of the place of use, the DC amplifier 2. It is generally susceptible to fluctuations due to external conditions such as fluctuations in the power supply to the pulse width modulator 3 and the pulse generator 4, which makes the output pulse width unstable and impairs the stability of the temperature converter.

In addition, in order to obtain a pulse width output that is correctly linearly proportional to the bridge output voltage, the DC amplifier 2 and pulse width modulator 3 are required to have sufficiently good linearity, but conventionally these points have been sufficiently resolved. It was hard to say that it was.

The present invention was made in order to eliminate the drawbacks of the conventional technology, and is that parts other than the temperature sensor are affected by external conditions (environmental temperature at the place where temperature is measured, fluctuations in power supply voltage, etc.). As a result, the output pulse width is not affected (any variation other than the pulse width fluctuating and taking a predetermined value according to the design conditions to express the temperature to be measured), and stable operation is achieved. The object of the present invention is to obtain a temperature measuring device that maintains good linearity between temperature and the time width of an output pulse expressing the temperature.

First Embodiment FIG. 2 is a circuit diagram showing a first embodiment of the present invention, a part of which is shown as a block diagram.

In FIG. 2, 1, 2, 3, 4. R1, R2, R3, and R4 each have the same meaning as the corresponding symbols in FIG. 1, and in addition, R5 is a fixed resistor, C1 is a capacitor, and 5 is a switching element, and in this example, a transistor was used. 6 is a frequency divider (frequency divider)
, 7 is a gate 8 is a counter, and a and b are symbols for pointing out locations on the circuit for explanation.

The pulse generator 4 generates a rectangular wave signal with a constant repetition rate f.

As an example of the pulse width modulator 3, for example, a known appropriate voltage-controlled one-shot multi-vibration brake is suitable.

A pulse with a constant pulse width (time width) is, period T, and peak value, which is the output of the pulse width modulator 3, is supplied between the base and emitter of the transistor 5.
As shown in Figure A, it is on only for time ts, and time t' (=T
-ts).

In order to explain this embodiment, FIG. 3 shows a part of FIG. 2 as an equivalent circuit.

In FIG. 3, the voltage between a and b of the bridge circuit is an integrated rectangular wave, and changes in the shape of a substantially triangular wave above and below the DC value Ec, as shown at the beginning of FIG.

In this case, if the time constant of capacitor C1 and resistor R5 is selected to be sufficiently larger than the pulse repetition period T, the voltage between a and b will be sufficiently smoothed to a nearly constant voltage E.
c.

Here, since the voltage between a and b is Ec, when the transistor 5 is turned on, the current i' flowing through the resistor R5 is 1=Ec/R5 (1).

Also, if the current flowing between a and b from the DC power supply 1 through the resistor R3 (considering excluding the directionality) is i, then the relationship 1-T=i'・ts is established, so 1= t @ The following relationship is obtained: T/ts (2J).

From equations (1) and (2), Ec/i = R5・T/l s (3) E
C is the voltage between a and b, and i is the current flowing between a and b, so in equation (3), Bc/i on the left side
represents the effective resistance between a and b, which can be expressed as R
If t, then Rt=R5·T/ts (4).

That is, the value of the practical resistance Rt between a and b is inversely proportional to the time 'ts during which the transistor 5 is on if the pulse repetition period T is constant.

In FIG. 2, when the bridge circuit is out of balance and a voltage Vi is output, this voltage is amplified by the DC amplifier 2 and input to the pulse width modulator 3, and the time width from the frequency divider 6 is input to the pulse width modulator 3. The pulse width of the pulse train at t is varied accordingly.

The pulse train subjected to pulse width modulation is input to the transistor 5, which controls the on/off operation of the transistor 5 to a-
The effective resistance value Rt between b is controlled.

Here, the pulse width modulator 3 is configured such that when the input voltage changes in the direction of decreasing (increasing), the pulse width ts of the output pulse train changes in the direction of increasing (decreasing). (decreases), the effective resistance Rt between a and b changes in the direction of decrease (increase) so as to satisfy the relationship of equation (4).

Now, when the conversion resistor R1 changes and its resistance value changes in the direction of increasing, the output voltage Vi of the bridge circuit changes in the direction of decreasing.

This output voltage Vi is amplified by a DC amplifier 2, inputted to a pulse width modulator 3, and changed to increase the output pulse width ts.

That is, when the conversion resistance R1 changes in the direction ζ in which it becomes larger, the effective resistance Rt between a and b in FIG. 3 changes in the direction in which it becomes smaller, so that the bridge circuit eventually reaches an equilibrium state at a certain point.

Similarly, when the conversion resistance R1 changes in the direction of decreasing, the pulse width ts changes in the direction of decreasing, and a
The effective resistance Rt between -b changes in the direction of increasing, and the bridge circuit becomes in a balanced state.

In this way, the bridge circuit always maintains a substantially balanced state and obtains a variable output such that the pulse width ts is determined according to the value of the conversion resistor R1.

If the polarity of the DC power supply of the bridge circuit is connected in the opposite direction to that shown in Figures 2 and 3, when the input voltage of the DC amplifier 2 changes in the direction of increasing, the pulse width modulator 3 Each part of the circuit is configured so that the pulse width ts of the output pulse train increases.

With this configuration, the bridge circuit is in a balanced state regardless of the direction of the DC power supply voltage.

In this balanced state, if the amplification degree of the DC amplifier 2 is sufficiently large, for example about 90 dB, the output Vi of the bridge circuit becomes very small, and the bridge circuit can be made completely balanced.

At this time, the following relationship holds true as is generally known.

R2・R3=R1・Ra That is, 1/Ra=R1/(R2・R3) (5) Here,
Ra is a combined resistance between bridge circuits a and 2 in FIG. 2, and the following relationship holds true.

1 / Ra = 1 / R4+1 / Rt = 1
/R4+(1/R5)(ts/T) (6) From the relationships in equations (5) and (6), the following relationship holds true when the bridge circuit is approximately balanced.

1/R4+(1/R5)・(ts/T)=R1/(
R2・R3) (7) If we calculate the pulse width ts of the output pulse of the pulse width modulator 3 using the relationship in equation (7), we get t 5 − T−R1・R5/ (R2・R3) − T−R5
/R4(8) That is, the pulse width ts is proportional to the conversion resistance R1.

Since the pulse width ts has a constant period T, the conversion resistance R1
It is related only to the change in the resistance value of , and is unrelated to the voltage E of the DC power supply 1 of the bridge circuit.

This pulse width ts expresses the temperature to be measured.

In FIG. 2, a signal of frequency f from a pulse generator 4 is divided into m by a frequency divider 6 and supplied to a pulse width modulator 3.

Since this divided signal has the previously explained period T and pulse width t, this period T is expressed as T=m/f (9).

On the other hand, the pulse (mts
The pulse is input to the gate 7 to open it, and the pulse from the pulse generator 4 is supplied to the counter 81 only during the period when the gate 7 is open, and the counter 8 counts the number of supplied pulses.

If the counter 8 counts the number of pulses n, then
Pulse width ts of the output signal of the pulse width modulator 3 at that time
is expressed as t s = n/f (10).

Substituting equations (9) and 00) into equation (8) and rearranging it, 02m-R1・R5/(R2・R3)-m-R5/R4
We get 0υ.

That is, the count value 11 of the counter 8 is proportional to the conversion resistance R1.

Further, the relationship of the Uυ equation is related only to the anonymous numbers m and n and the resistance values R1 to R5, and is unrelated to voltage fluctuations and frequency fluctuations.

For example, if the counter 8 is a decimal counter, the conversion resistance R
The measurement result based on 1 can be understood as a decimal number, and if a binary counter is used, the measurement value can be understood as a binary number.

If you want to further improve the precision of the measurement H frequency, for example, by increasing it by a factor of q, increase the division ratio of the frequency divider 6 and further increase it by a factor of q (the division ratio is m-q times), and send it to the pulse width modulator 3. If the period of the input pulse train is multiplied by q (the pulse width remains t), the count value obtained by the counter 8 will be multiplied by q.
), it is possible to measure with an accuracy of q times.

Note that the DC amplifier 2 in FIG. 2 is not essential, and if the output voltage of the bridge circuit is sufficient, its output may be directly applied to the pulse width modulator 3.

In the second embodiment shown in FIG. 5, 21 is an operational amplifier, and 22 is an impedance for feedback, and both of the symbols 21 and 22 correspond to the DC amplifier 2 in FIG.
It is surrounded by a dotted chain line and indicated by the symbol 2.

), and the part 3 surrounded by the one-dot chain line is a functional part corresponding to the pulse width modulator 3 in FIG. 2, and is obtained by an IC chip, for example.
32 is a comparator, 33 is a flip-flop, R41R42, R43 are three resistors with the same resistance value, Tr is a transistor, 34, 35, 36, 37 are each a signal input terminal, 38 is a signal output terminal, 39 is a power supply It is a terminal.

Further, 10Eb is the positive terminal of the power supply, Rt and Ct are resistors and capacitors for setting the time constant, and other symbols have the same meanings as the corresponding symbols in FIG.

Equation (8) holds true for FIG.

FIG. 5 shows the frequency divider (divider) 6 of FIG. Although the gate 7 and counter 8 are not shown, the equation 01) can be adapted and implemented by adding these as necessary to create a block configuration similar to that shown in FIG.

Moreover, FIG. 5 shows how to insert the DC type 81 and the output voltage Vi.
Although the way in which the resistors R1, R2, R3 and R4 are combined is different from that shown in FIG. 2, there is no problem due to this.

The operational amplifier 21 may be of a known type, and the common connection point of resistors R2R4 and R5 is connected to the negative input, the connection point of the resistors R1 and R3 is connected to the positive input, and the output voltage is connected to the negative input via the feedback impedance 22. When the feedback is returned to The capacitor C1 is not required in this case.

Here, if a capacitive element is used for the feedback impedance 22, the circuit 2 corresponding to the DC amplifier can be provided with an integrating function.

The output is taken out from one side of the flip-flop 33, and the same signal is fed back to the transistor 5 of the bridge circuit to control its switching operation.

FIG. 6 shows a waveform diagram when the circuit 2, which corresponds to a DC amplifier, is provided with an integration function, and in this diagram, A is the pulse generator 4.
A pulse train of constant period T and pulse width t sent from B
C is the output waveform of the operational amplifier 21 of the circuit 2, which corresponds to a DC amplifier, and C is the waveform applied to the base of the transistor 5.

The voltage at the connection point of the resistors R2, R4 and R5 (the negative input of the operational amplifier 21) of the bridge circuit in FIG.
The voltage is set to be slightly higher when the transistor 5 is off and slightly lower when the transistor 5 is on than the voltage at the connection point of the resistor R3 and the resistor R3.

In the [C chip 3 in FIG. A discharge current flows from
A pulse with a time width ts is output from the terminal 38.

The comparator 32 determines the trigger level, and compares the voltage at the junction of the resistors R42 and R43 with the voltage of the trigger signal from the terminal 36, and opens the flip-flop 33 when the trigger signal voltage is higher. Reset.

The output of the operational amplifier 21 has the waveform shown in FIG. 6B,
When the conversion resistance R1 changes, the integrated and averaged level of the operational amplifier 21 (indicated by symbol Eth in FIG. 7B) changes.

This is applied to the terminal 34 as a control input of the IC chip 3 of the pulse width modulator-equivalent functional circuit, thereby providing it as an input to the comparator 31.

The comparator 31 compares this input voltage with a set reference potential at the connection point between the resistor Rt and the capacitor Ct, and outputs an output to the comparator 31 when the voltage at the connection point between the resistor Rt and the capacitor Ct is higher. and set the flip-flop 33.

When the flip-flop 33 is set, the transistor T
r is turned on and the charge in the capacitor Ct is rapidly discharged.

When a trigger signal is input from the terminal 36, the comparator 32
operates to reset the flip-flop 33 and turn off the transistor Tr.

From the moment the transistor Tr is turned off, the capacitor Ct is charged toward +Eb with a time constant determined by the resistor Rt and the capacitor C1.

When this voltage exceeds the control voltage input from the terminal 34, the flip-flop 33 is turned on and the same operation is repeated thereafter.

In this case, regarding the reference potential set at the connection point of the resistor Rt and the capacitor Ct, even if this reference potential fluctuates somewhat, there is no problem as long as the fluctuation period is sufficiently thicker than the period T of the pulse T, so the capacitor Ct is charged. Even if the current to be output is not a constant value, that is, even if a constant current circuit is not used in this part, the output voltage of the operational amplifier 21 is increased until the pulse width ts of the output pulse from the terminal 38 becomes a value proportional to the conversion resistor R1. The entire circuit of FIG. 5 is balanced so that the average level of , ie, the control voltage at terminal 34, changes automatically.

Further, the time constant of the resistor Rt and the capacitor Ct is set to be somewhat smaller than the period of the pulse from the pulse generator 4, so that the time width ts of the output signal pulse reaches the period T of the pulse from the pulse generator 4. I'll make sure this doesn't happen.

Father, when using a transistor as a switching element,
The present invention can be practiced using either grounding method.

For example, in the case of a device mounted on a flying object such as a radiosonde to measure temperature, it is necessary to send the measurement result as a modulated signal using radio waves to receiving equipment on the ground, and the usage conditions of the radiosonde itself are Due to the constraints imposed by the It meets the requirements of the purpose of use.

[Brief explanation of the drawing]

Fig. 1 shows a conventional example, a circuit diagram partially shown as a block diagram, Fig. 2 a circuit diagram partially showing a first embodiment of the present invention as a block diagram, and Fig. 3 a circuit diagram partially shown as a block diagram. Regarding the embodiment shown in FIG. 2, the main parts of the present invention are shown as circuits, FIG. 4 is a waveform diagram for explaining the embodiment shown in FIG. 2, and FIG. FIG. 6 is a circuit diagram partially showing the second embodiment as a block diagram, and FIG. 6 is a waveform diagram for explaining the embodiment shown in FIG. In each figure, the main symbols are as follows. DESCRIPTION OF SYMBOLS 1... DC power supply, 2... DC amplifier, 3... Pulse width modulator, 4... Pulse generator, 5... Transistor, 6... Frequency divider, 7... Gate , 8...
・Counter, R1-R4...Bridge circuit resistance, 21
...Operation amplifier, 22...Feedback impedance, 3
1, 32...Comparator, 33...Flip-flop, Rt...Resistance for time constant, Ct...Capacitor for time constant.

Claims (1)

[Claims] 1. A bridge circuit is provided on one side of the conversion resistor whose resistance value is determined in response to temperature, and a switching circuit is provided on the other side, and a smoothing circuit is provided to smooth the intermittent output from the switching circuit. In a temperature measuring instrument that performs pulse width modulation in response to the output of the bridge circuit and negatively feeds back the pulse width modulated pulse to the switching ratio control input of the switching circuit, the frequency-divided pulse obtained by dividing the clock pulse is Output of the bridge circuit; thus, pulse width modulation is performed, and the pulse width modulated frequency divided pulse is negatively fed back to the switching ratio control input of the switching circuit, and the above pulse width modulated frequency division pulse that appears during the existence time of the pulse width modulated frequency divided pulse is A temperature measuring device that counts the number of clock pulses and calculates the temperature based on the above counted value.