JPH0346771B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a pressure transducer, and more particularly to a pressure transducer that takes into consideration a method of compensating the temperature of an element and a method of coupling it with a peripheral circuit.

In a pressure detection device in which a semiconductor diffused resistance layer is used as a gauge resistor and a bridge circuit is constructed using this gauge resistor, compensation for temperature fluctuations is an important issue. For example, as shown in Fig. 1, gauge resistance R ₁ exhibiting resistance changes of opposite signs in response to pressure,
In a conventional full bridge circuit configured using R ₂ and R ₃ and R ₄ , compensation resistors rs ₁ and rs ₂ are inserted in series with gauge resistors R ₁ and R ₂ , respectively, and compensation resistors are inserted in parallel. Bridge zero point compensation is performed by connecting rP ₁ and rP ₂ respectively. By appropriately setting the resistance values of these compensation resistors rs ₁ , rs ₂ , rP ₁ , and rP ₂ , the apparent temperature coefficients of the gauge resistances R ₁ and R ₂ can be determined within the desired temperature range. Resistance R ₃
can be made the same as the temperature coefficient of _R4 , and we are trying to perform zero point compensation for temperature.

However, the above conventional zero point temperature compensation has the following major problems. In other words, the pressure sensitivity of the gauge resistance cannot be compensated for only by the above-mentioned zero point temperature compensation.
For example, as shown in Figure 2, the bridge drive voltage V _B
It became necessary to compensate for pressure sensitivity by varying the This pressure sensitivity compensation is performed using a temperature-sensitive resistance element such as a thermistor.
A temperature-dependent voltage V _B of the reference voltage V _E via RT
as a bridge or as an amplifier.
The purpose is to vary the bridge drive voltage _VB by setting it variably using the gain of the amplifier and the temperature-sensitive resistor RT, thereby providing temperature compensation that matches the pressure sensitivity temperature characteristics of the gauge resistors _R1 to _R4 . It is something to do. However, when this temperature compensation is performed, since the drive voltage V _B changes depending on the temperature, the pressure sensitivity compensation and the zero point compensation change at the same time, and as a result, it is very difficult to make an actual adjustment, and it is difficult to make an optimal adjustment. was almost impossible. A second problem is that since a fixed resistor is connected in parallel with the gauge resistor, the linearity of the bridge output decreases, making highly sensitive detection impossible.

Therefore, the present inventors have provided a new method for zero-point temperature compensation when applied to the half-bridge circuit. (Japanese Patent Application No. 54-126552) This compensation method is based on the following principle.

The value of each strain resistance assembled in the bridge,
If the temperature coefficients were exactly equal, changes due to temperature should cancel each other out. However, in reality p
The temperature coefficient of the shaped diffusion layer is 3000~ even at room temperature.
It is extremely large at 3500 ppm/°C, and the temperature coefficient itself changes with higher-order terms related to temperature.

Since the change in semiconductor diffused resistance with respect to temperature is quadratic, the change in zero point with respect to temperature exhibits a complicated curved pattern.

As shown in FIG. 3, a fixed resistor having no temperature coefficient is inserted between the strain resistors of the bridge circuit. If we trace the potentials (offset voltages △V H , △V L ) at points H and L (offset voltages △V _H , △V _L ) seen from the 1/2 point on the dummy resistor side of the half bridge from a low temperature of -20°C to a high temperature of 80°C, we can see Figure 4. As schematically shown in , △V _H and △V _L
There is a state in which the curvature of temperature fluctuation is reversed in a certain state ( _ΔV As shown in Figure 2, in the method of compensating temperature fluctuations in pressure sensitivity by moving the bridge applied voltage, the temperature (T) and the bridge drive voltage (V _BT ) correspond in a proportional relationship. . FIG. 5 shows potential changes at each point of the bridge in FIG. 3 based on bridge drive voltage instead of temperature.

The midpoint potential on the dummy side of the bridge is a straight line of 1/2V _BT . If the midpoint of the fixed resistance between the strain resistances is X, then between the △V _H and △V _L curves there are
There should be a straight line (broken line △V _x ) passing through the three high temperature points. The slope of the midpoint on the dummy side changes by 1/2, so if we assume a point Y, there must be values of X and Y that make the broken line of △V _X parallel to YV _BT (two-point broken line). It is. At this time, the potential difference δV between point X and point Y remains constant regardless of changes in V _BT (temperature changes).

This relationship shows that the potentials at points H and L are 2 with respect to V _BT .
This can be explained by approximating with the following equation. Now, the bridge drive voltages at low temperature, medium temperature, and high temperature are V _BT (L),
V _BT (C), V _BT (H), the potential difference between the dummy side 1/2 point and each point H and L, respectively, are △V _H (L), △V _H (C), △
Let V _H (H), △V _L (L), △V _L (C), and △V _L (H). An arbitrary point on the resistor inserted on the strain resistance side and an arbitrary point on the dummy low resistance side are taken as a ratio, and are designated as X and Y, respectively. Potential at points H and L seen from the ground point △V _H
(V _BT ), △V _L (V _BT ) are, △V _H (V _BT ) - 1/2V _BT = AV ² _BT + (B - 1/2) V _BT +C = △V _H (L), △ V _H (C), △V _H (H)
...(1) △V _L (V _BT ) - 1/2V _BT = aV ² _BT + (b - 1/2) V _B +C = △V _L (L), △V _L (C), △V _L ( H)
...(2) Here A, B, C, a, b, c are
V is a coefficient of a quadratic equation regarding _BT .

If the difference between △V _H and △V _L is δV, then from (1)-(2), δV (V _BT ) = [(A-a)X+a] (V _BT ) ² [(B-
b)X+(b-Y)] _VBT +(C-c)X ^+C ...(3). In order for δV(V _BT ) to be constant regardless of V _BT , the second-order and first-order terms regarding V _BT need to become zero. The conditions that satisfy this are: X=-a/(A-a), Y=(Ab-aB)/(A-a)
, V=(Ac-Ca)/(A-a)...(4). By satisfying this condition, the offset voltage ΔV becomes constant regardless of _VBT , and a compensation condition that does not depend on the external temperature can be found.

With the method described in detail above, extremely highly accurate zero point temperature compensation is possible. However, the biggest drawback of the conventional method is that the temperature compensated bridge is limited to a half bridge. Needless to say, a half bridge has half the output as a full bridge, and from the perspective of the amplification circuit, it is also a loss in terms of signal-to-noise ratio. In addition, in order to assemble strain gauges diffused on the same semiconductor substrate into a bridge,
Different reverse bias voltages are applied to the substrate and the individual strain gauges. Even if strain gauges are diffused in the same substrate, if the reverse bias voltage is different, the temperature coefficient of resistance will be different, resulting in zero point drift depending on the reverse bias voltage. In comparison, the full bridge method in which half bridges are placed opposite each other not only doubles the output, but also has the advantage that the above-mentioned dependence of the zero point temperature drift on the reverse bias voltage is canceled out.

The object of the present invention is to provide a new converter using a zero-point temperature drift compensation method that completely overcomes these drawbacks.

First, explanation will be given starting from FIG. FIG. 6 shows a circuit configuration when determining the x and y conditions explained in FIG. 3 for a full bridge. Let the conditions for the left strain gauge be x, y, and the conditions for the right strain gauge be x', y'. The dummy resistor is only needed to determine x and x', and will eventually become unnecessary. Now, x, y, x′, found in the circuit of Figure 3,
The potential (as seen from ground) at each point corresponding to y′ is △
If V _x , △V′ _x , and the potentials under the condition that the potentials are constant when viewed from the dummy are δV and δV′, the result will be as shown in FIG. As you can see, the output of Full Bridge,
In other words, the output between x and x′ is △V _x −△V′ _x ,
It is proportional to V _BT . Figure 8 is a redrawn version of this to make it easier to understand. △V _x −△V′ _x consists of a component ZV _BT proportional to the voltage applied to the bridge and a constant potential δV _xx ′ component. The two components thus generated can be removed by the circuit configuration described above.

In FIG. 9, gauge resistances R _p , R' _p , R _o , R' _o that show resistance increases and decreases according to pressure are configured in a full bridge shape 81 , and a power source that changes according to temperature.
Driven by V _BT . Also full bridge 81
Fixed resistors R ₁ , R ₂ , variable resistor R ₃ , and power supply ΔV _xx ' are connected in series in parallel with . The output of the bridge is determined by the ratio x of the variable resistors r ₁ and r ₂ inserted under the condition that it shows a linear change with respect to the change in the power supply V _BT ,
It is determined by x′. One of the outputs is coupled to capacitor _C2 and the other to capacitor _C1 via analog switch _SA3 . The output terminal of the variable resistor R ₃ is selected to be a variable component ZV _BT in the output of the bridge and is connected via an analog switch SA ₁ to a capacitor C ₁ and then to a capacitor C ₂ . Also, variable resistor R ₃ is coupled in series with fixed resistor R ₂ and constant voltage δV _xx ', and together with one end of the bridge, an analog switch is connected.
Located in SA ₃ . analog switch here
SA ₁ and SA ₂ are opened and closed in the same phase by the drive pulse P, and SA ₃ is inverted and driven by the inverter I. The operation is performed in two shifts as follows.

(1) When SA ₁ and SA ₂ are on and SA ₃ is off, the potential ZV _BT + from the output terminal of variable resistor R ₃
δV _xx ′ is charged to the capacitor C ₁ .

(2) When SA ₁ and SA ₂ are off and SA ₃ is on, the composite value of pressure output and temperature drift component is superimposed on the charge charged to C ₁ from the output terminal of bridge 81 and sent to C ₂ . Charged.

Here, C ₁ has already been charged with ZV _BT + δV _xx ' due to 1), but since this value is a temperature drift component of the output of the bridge 81 , they cancel each other out, and C ₂ has only the pressure output. is charged. That is, the pressure output from which temperature drift has been eliminated appears as an external output.

The example detailed above is a case where a fixed resistor is inserted into the bridge circuit to actively compensate for zero point temperature fluctuations. In order to obtain this compensation condition, x and x' must be determined by actively exposing the element to low, medium, and high temperature conditions. Instead, once the conditions are determined, high accuracy of 0.1% full scale or higher can be obtained. Apart from such high-precision compensation, it can be used to reduce costs if only a few percent of the full scale can be adjusted, and only room temperature can be adjusted. In this case, a fixed resistor is not inserted into the bridge circuit, and an internally connected element is manufactured from the beginning. Similar to the manufacturing method for ICs, this method uses automatic inspection equipment to select only those that pass the standards. The present invention is also effective in canceling out the bridge offset voltage in such cases. FIG. 10 shows a circuit configuration showing an example thereof.

In this figure, the strain gauge R _p that feels pressure,
No fixed resistor is inserted in the full bridge circuit 91 that combines R' _p , R _o , and R' _o . The bridge is driven by a temperature-dependent power supply V _BT , and a bleeder resistor 92 is connected in parallel with the power supply V BT in which fixed resistors R ₁ , R ₂ and variable resistor R ₃ are connected in series. Assume now that the output of the bridge circuit 91 is the pressure output ΔP(P) and the zero point output (offset voltage) δV _p superimposed. The variable resistance R ₃ of the bleeder 92 is compared to ω
and set R ₃ ×ω×V _BT = δV _p . In this case, the interlocking switches SM ₁ and SM ₂ close at the bottom. Analog switch is activated by clock pulse P.
When SA ₁ , SA ₂ , SA ₅ , and SA ₇ are turned on, the variable resistance
The δV _p voltage from R ₃ is charged to capacitor C ₁ . In this case, analog switch SA ₃ ,
Since the pulse waveforms of SA ₄ and SA ₆ are inverted by the inverter I, they are in an off state. Next, SA ₁ , SA ₂ ,
When SA ₅ and SA ₇ are turned off and SA ₃ , SA ₄ and SA ₆ are turned on, the output ΔP+δV _p of the bridge 91 is superimposed on the previously charged potential of C ₁ . At this time, since δV _p previously charged to C ₁ and ΔP+δV _p of the bridge output have different polarities, -δV _p +ΔP+δV _p →ΔP. In other words, the constant potential component of the pressure output can be canceled out. The ΔP produced here is amplified by the differential amplifier _OP1 , and then again amplified by the next stage operational amplifier _OP2 .

Here, the polarity of the offset voltage component δV _p is different,
In the case of ΔP-δV _p , the same canceling effect can be obtained by turning the interlocking switches SM ₁ and SM ₂ of FIG. 9 upward and reversing the polarity of the output from the bleeder 92 .

As described in detail above, by using the method according to the present invention, it is possible to eliminate the offset voltage caused by whether or not a fixed resistor is inserted into the bridge circuit for pressure detection. In the conventional method, when the polarity of the offset voltage is different, it is necessary to use two power supplies, but the present invention has the advantage that only one power supply is required.

[Brief explanation of drawings]

FIGS. 1 and 2 are block diagrams showing examples of conventional devices, respectively; FIG. 3 is a bridge block diagram for explaining the half-bridge zero point compensation method already proposed by the inventors; and FIGS. 4 and 5. 6 is an explanatory diagram of the compensation operation, FIG. 6 is a configuration diagram when FIG. 3 is extended to a full bridge, FIGS. 7 and 8 are explanatory diagrams of the operation of a full bridge, and FIG. 9 is a diagram of the zero point temperature according to the present invention in a full bridge. FIG. 10, which is an example of a circuit configuration to which the compensation method is applied, is another configuration example of the present invention.

Claims (1)

[Claims] 1. A full bridge circuit that has a pair of drive voltage application terminals and a pair of output terminals, and is configured by connecting a pair of gauge resistor pieces in parallel that exhibit resistance changes of opposite signs in response to pressure, and the drive of this full bridge circuit. A power supply circuit that variably sets and applies a drive voltage between the voltage application terminals in response to temperature changes, and a connection between the drive voltage application terminals of the full bridge circuit so that any potential from the drive voltage to the ground potential can be obtained. a bleeder resistor consisting of a variable resistor and a plurality of fixed resistors, a first capacitor for charging the potential acquired by the bleeder resistor, and a potential acquired from the bleeder resistor across both ends of the first capacitor. a second capacitor connected between one output terminal of the full bridge circuit and one end of the first capacitor; A third analog switch connected between the other output terminal and the other end of the first capacitor and opened and closed in a complementary manner to the first and second analog switches. pressure detection device.