JPS61200457A - Air/fuel ratio detector - Google Patents

Abstract

PURPOSE:To enhance detection accuracy, by compensating temp. on the basis of information relating to the temp. of a solid electrolyte present in the output signal of an air/fuel ratio detector so that the output of the air/fuel ratio detector receives not effect of circumferential temp. CONSTITUTION:When switches 6a, 6b are turned ON and switches 7a, 7b are turned OFF, a current Is flows to a solid electrolyte to perform sensing opera tion. A sample hold circuit 9 holds terminal voltage VH after the time td from the start of sensing operation by a delay circuit 8. The held VH is compared with a certain constant voltage VHref by a comparator 14 and, when VH<VHref, an ON-signal is inputted to the base of a transistor 15 and, when VH>ref, an OFF-signal is outputted to the base of the transistor 15. That is, when the temp. of the solid electrolyte 1 become lower than set temp., a current is sup plied to a heater 3. Because VH can be controlled to a certain value VHref by this method, the temp. of the solid electrolyte 1 is always controlled to a con stant value and the detection of an air/fuel ratio is enabled regardless of the effect of circumferential temp.

Description

Detailed Description of the Invention [Field of Application of the Invention] The present invention relates to a sudden fuel ratio detector for closed-loop control of the air-fuel ratio of an internal combustion engine, and in particular to a sudden fuel ratio detector suitable for obtaining an output that is not affected by ambient temperature. Concerning vessels.

[Background of the invention]

As described in JP-A-57-192852, conventional devices use alternating current for the solid electrolyte, provide a temperature measurement period, or provide a temperature measurement element as a means for detecting the temperature of the solid electrolyte. It became. It is also known that the temperature of the electrolyte can be estimated from the exhaust gas temperature and the operating condition of the engine (Unexamined Japanese Patent Publication No. 59-103265, Unexamined Japanese Patent Publication No. 59-188054). There were problems such as insufficient accuracy due to the manual temperature detection.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide a sudden fuel ratio detector that obtains an air-fuel ratio signal that is not affected by ambient temperature.

[Summary of the invention] ゛The present invention has discovered that information related to the temperature of the solid electrolyte is contained in the output signal of a sudden fuel ratio detector,
Based on this information, temperature compensation is performed so that the output of the sudden fuel ratio detector is not affected by the ambient temperature.

[Embodiments of the invention]

Prior to the description of the present invention, one principle underlying the present invention will be explained with reference to FIGS. 1 to 6M. A porous diffusion resistor 2 is provided on the exhaust side of the solid electrolyte 1 . The solid electrolyte 1 is shaped like a bag tube, and the atmosphere is guided inside. Furthermore, a heater 3 is built in inside. The solid electrolyte Jgt1 is provided with electrodes 4a and 4b on the exhaust side and the atmosphere side, respectively, and uses a constant current circuit to
Apply constant forward and reverse currents to WJ@pole in a time-division manner.

The output is obtained from the change in the terminal voltage V at this time. Figure 2 shows
One sensor is an enlarged view of the area marked with a circle in Figure 1.

In order to also measure the rich air-fuel ratio, the direction of the current applied to the solid electrolyte 1 is reversed in a time-sharing manner. As shown in Fig. 2, the operation is as shown in Fig. 2. First, I flows in the direction of the arrow in the figure, and the a element in the atmosphere flows into the diffusion resistor 2 in the exhaust gas.
Next, oxygen is caused to flow in the opposite direction to that of I, and oxygen is extracted from the inside of the diffusion resistor 2.

The former is called a bias operation, and the latter is called a sensing operation. This bias operation makes it possible to measure the rich air-fuel ratio.

FIG. 3 shows changes in the distribution of oxygen concentration within the diffused resistor 2 (actual curve 11A in the figure) and changes in the terminal voltage V of the solid electrolyte [1]. FIG. 3(a) is a diagram during bias operation.

When oxygen is sent into the diffused resistor 2 by flowing the gas, the oxygen concentration distribution inside the diffused resistor 2 becomes higher on the solid electrolyte 1 side than the oxygen concentration P in the exhaust gas on the exhaust side, and eventually a certain distribution curve is formed. converges to. Therefore, the terminal voltage V also increases as the oxygen concentration of the electrode 4a increases, and eventually converges to a constant value. V at this time is.

4F P(4a) where r: Internal resistance of solid electrolyte 1 T: Temperature of solid electrolyte 1 R: Gas constant F: Faraday constant P(4a): Oxygen concentration on electrode 4a side P(4b): Electrode 4b side The second term on the right side is the electromotive force, and as this value decreases, V increases. This bias operation creates a state in which P (4a) is higher than P, making it possible to measure rich air/fuel conditions.

Figures 3(b) and 3(c) are during sensing operation, and (
b) is a lean air-fuel ratio, and (C) is a rich air-fuel ratio. The terminal voltage V at this time is.

It is expressed as In FIG. 3(b), when the oxygen in the diffusion resistor 1 is drawn out by flowing L, the acid lRa on the electrode 4a side
The degree gradually decreases (solid straight line in the figure) and eventually reaches zero. At this time, the electromotive force of the second term on the right side of equation (2) is superimposed on the terminal voltage V. The increase due to this electromotive force is a constant value E.

If the sensing operation is terminated when t is reached, the time t at this time becomes a value proportional to the air-fuel ratio.

Figure 3(c) shows the case of a rich air-fuel ratio, and the oxygen concentration distribution after bias operation is due to the reaction between oxygen and combustible gases (Go, HC, Hl) diffusing from the exhaust side. =
A region of 0 occurs. Therefore, since the amount of oxygen detected during sensing decreases, the time t1 until the change in ■ becomes E, 1 is:

A rich air-fuel ratio can be measured by 1 or more, which is smaller than 1.

FIG. 4(a) shows 1. , L, and V timing charts are shown. As shown in FIG. 4(a), 1. , I, flow in a time division manner while reversing the direction a 1% + Xm
is a constant current, and the bias operation time t is constant,
during sensing operation.

The measurement is ended when the increase in (2) reaches E-L (constant M), and the air-fuel ratio can be determined by measuring time t1 at this time. Figure 4(b) shows the output characteristics of one hydraulic formula, t
, is proportional to the air-fuel ratio over a wide range from rich to lean.

FIG. 5 shows an embodiment of a circuit for carrying out the operation of a single sensor. 5 is a constant current source, and when switches 6a and b are turned ON and switches 7a and 7b are turned OFF, current flows into the solid electrolyte 1 and performs a sensing operation.
When a and b are set to 0FFL, and switches 7a and b are turned on, the current flows and bias operation is performed, here I- and "Is." Hereinafter, the operation of this circuit will be explained using the waveforms shown in Fig. 6(a). First, we will explain the point at which the ON signal is input to the switches 6a and Eib and the sensing operation ((a) in FIG. 6(a)) starts.This ON signal is input to the delay circuit 8. Then, as shown in FIGS. 6(a)-(a), an ON signal is applied to the sample and hold circuit 9 after t4.The sample and hold circuit 9 receives the terminal voltage V after t4 from the start of the sensing operation. , hold.

Em& is added to this V by an adder circuit 10 and inputted to a comparator 11. In comparator 11, ■
.

A trigger signal is output when the terminal voltage V becomes larger than +Eo. This trigger signal is input to the monostable multivibrator 12, and between the trigger signal input device t2, the end is turned off.
Turn FF and Q on.

That is, during this period 31, an ON signal is input to switches 7a and 7b, and an OFF signal is input to switches 6a and 6b, and a bias operation is performed (Fig. 6 (a) to (b)).
When the elapsed time has elapsed, the Q of the multivibrator 12 is turned ON again.
Since Q is in the OFF state, sensing operation is started.

The output is produced by an output circuit 13 that converts the time when Q is ONL, that is, tl, into an analog output.

Here, the held V is almost equal to rI shown in equations (1) and (2), and indicates the internal resistance r of the solid electrolyte 1 (,'1.=X,=constant), and also indicates the temperature of the solid electrolyte 1.

Next, the temperature T of [lI is T,'(T,>T,'')
The change in the waveform in the case of the change will be explained with reference to FIG. Figures 6(a)-(a) are waveforms during sensing,
(a) to (b) are waveforms during bias. The temperature is T,
When it becomes low from T,', ■.

becomes high as V, / (formulas (1) and (2)). Here, if El is the same in both cases, the sensing time is longer when T,', and ' is shorter. Note 1 Figure 6
The bias waveforms (a) to (b) do not change because tb is kept constant, and the absolute values are still T, /
It becomes higher when t, and an error occurs in t depending on the temperature. The reason why an error occurs in t in this way is explained in Figure 6 (b
). In FIG. 6(b), one distribution showing the oxygen concentration distribution in the diffused resistor 2, (distrib
tion in 5ensin (action).

The distribution at the end of the sensing operation is 1 distribution D1 (distr
ibution in bias action)
indicates the distribution at the end of the bias operation *Da'
+Db' is.

s showing the respective distributions when the temperature is T,'
If Ib+L+tb+EaL are set to fixed values, there will be differences between D and D@' and between Db and Db'. This is because the diffusion rate of the v element within the diffused resistor 2 differs due to the difference in temperature. here.

Since D,,D,' holds E and L constant, W1
Oxygen is extracted until P(4a) at the cap 4a becomes almost zero, resulting in almost the same distribution. on the other hand.

Since tb is constant, DbtDb' is different due to the difference in oxygen diffusion rate, and D5' is more Dbt than D5'.
l has a higher overall distribution.1 As shown above, when the ambient temperature changes, the change in the oxygen concentration distribution due to the oxygen diffusion rate is different, so there is a change in t. arise. Since t,′ becomes smaller than t, the sixth
As shown in Figure (c), the output at T,' is T.
, and an error will occur depending on the temperature.

FIG. 7 shows an embodiment of a circuit for preventing errors caused by the above-mentioned temperature changes. The V held by the sample and hold circuit 9 (value indicating internal resistance) is compared with a certain value V 31 t a 1 by the comparator 14, and when VW<V, . . . , an ON signal is sent to the transistor 1
5, the transistor 15 is turned off, and the power supply to the heater 3 is stopped. This V*<
V − and -t are states in which the internal resistance r of the solid electrolyte 1 is small, that is, the temperature is higher than the set temperature, and in this case, as described above, the power supply to the heater 3 is stopped.

Next, when V, > V, , the comparator 14 becomes OF
In order to output the F signal to the base of the transistor 15, the heater 3 is energized 6, that is, when the temperature of the solid electrolyte 1 becomes lower than the set temperature.

When the heater 3 is energized by 0 or more, ■ becomes a certain value V
Since the temperature of the solid electrolyte 1 can be controlled to @t*t, the temperature of the solid electrolyte 1 is always controlled to a constant value. This makes it possible to detect the air-fuel ratio regardless of the influence of ambient temperature.

Figure 8 shows the operating principle and experimental results of the circuit in Figure 7.Since the temperature of the solid electrolyte 1 is kept constant, the waveforms during sensing in Figures 8(a)-(a) also (a)-(
The waveforms at the time of bias in (b) are also the same at T, and T,'. Also, as shown in Figure 8(b), ri, tori, ', D
, D, and I are the same.The output value measured using the principle of 0 or more is shown in Figure 8(c).Even if the ambient temperature is different from T,,T,', the output value of heater 3 is This is because the temperature of the solid electrolyte 1 is kept constant through control.

The outputs when TstTa2 match.

Next, another temperature compensation principle is shown in FIG. 9, where X,
=1. = constant, El constant. Figure 9(a)-(b)
As shown in , the value V@H held during sensing
V11' is maintained until biasing (Fig. 9(a)-(b)), and the terminal voltage during biasing is Vs 75<V.
When it becomes equal to W, the bias operation is terminated. Therefore, when T,' and 8g1a degrees are low, t,
′ and the bias time becomes longer, so the sensing time t
There is no difference between , and t, /. This principle is illustrated in Figure 9 (
Shown in b).

To make the terminal voltage V at the end of bias equal to VW (terminal voltage at the start of sensing), formula (1) and (2
) is to make V in the equation equal, which is.

P (4a) = P (4b)
...This occurs when (3) occurs. In other words, P (4
Since a) is the oxygen concentration in the atmosphere, P(4a) is
The concentration of oxygen increases until it reaches the level of oxygen in the atmosphere. In other words.

Since P(4a) always takes a constant value regardless of the ambient temperature (T,,T,'), D,' and D5 have almost the same distribution. Therefore, as shown in Figure 6(b),
The difference between and p% disappears, and t, which is unaffected by temperature, is obtained. Figure 9(c) shows the actual measurement results, and as the ambient temperature changes from T, to T,', the output time becomes longer, which is not affected by the ambient temperature. It will be done.

FIG. 10 is an embodiment of a circuit for carrying out the operating principle of FIG. 9. V held and held by the sample-and-hold circuit 9 is input to a comparator 16 and compared with the terminal voltage V when I is flowing during bias.

When , and V become equal, the comparator 16 turns OF
Fi: Outputs the number and this is monostable multivibrator 1
Input to the reset terminal of 2, -ζ- is ON, Q is OFF
F, switches 6a, b are ON, switches 7a, b
turns off. Then, ■ is grounded, the comparator 16 is turned ON, and the multivibrator 12 is QON,
When the amount of change in the terminal voltage V becomes larger than IEat during the zero-order sensing in which the QOFF state continues, the comparator 17 turns OFF. This OFF? No. m is input to the preset terminal of the multivibrator 12, and OK is O.
FF and Q are turned on and bias operation begins. At this time v
1 is grounded, so comparator 17 immediately turns ON.
,, the multivibrator 12 is LOW OFF, QON
The circuit of FIG. 10 executes the compensation operation shown in FIG. 9, and automatically obtains a temperature-compensated output.

As described above, it can be seen that when the temperature of the solid electrolyte 1 is low, it is sufficient to lengthen the bias time. fjS9.
In the method shown in Figure 10, t changes automatically, but in the method shown in Figures 11 and 12 below.

This method forcibly changes t. In other words, as shown in Fig. 11, when the temperature changes from T, to T,', the change in y,,y,' is detected, and in accordance with this change, t, is lengthened by 1, /. This allows the bias time to be lengthened and the output to be kept constant. Here is an example of a circuit that realizes this operation.

As shown in FIG. 12, vl held by the sample hold circuit 9 is taken into the microcomputer 19 via the A/D converter 18. , in the microcomputer 19, outputs a signal that is ON only for a predetermined bias time t2 based on V. This ON signal turns on switches 7a and 7b.
Switches ea and 6b are turned off by the action of the inverter 20, and the bias operation is connected only during t5.

When the bias time t has elapsed, the output from the computer 19 is 0FFL/, and the switches 7a and 7b are turned off.

On the other hand, due to the action of inverter 2o.

The switches 6a and 6b are turned on to start sensing operation. When the output voltage v,l increases to v1', the bias time tb is outputted as long as th'.

Note that within the microcomputer 19.

Rotation speed signal 21. Load signal 22. Intake air amount 23° Cooling water @24. Intake temperature 25. The bias time t1 may be corrected depending on the exhaust gas temperature 26.

13 and 14 show other temperature compensation circuit terminals.
In this configuration, when the degree of one circumference changes from T, to T,', the bias time t, remains unchanged, and the current value 1. It changes the In other words, the amount of change in T.

When the zero-peripheral degree T1, which is detected by the amount of change in VW and changes the bias current, is as low as 7.1, the bias current 1. , I and large shi (Fig. 13(b)
)-C port)), even with the same bias time.

Make sure the bias oxygen amount is the same. This results in
It is as if the bias time was lengthened, and the temperature dependence of the output disappears. In addition, Fig. 13(b)-(a)
The current during sensing is the same as shown in .

FIG. 14 is basically the same as the circuit N configuration in FIG. 5, but with the power supply 5 of the current generator during sensing.

The difference is that the power supply for the bias current is separate. In FIG. 14, the bias current circuit is made as follows. That is, the OP amplifier 27 and the transistor 2
8, the voltage across the resistor R1 is controlled to be v8. For this reason, the bias current engineer is ■.

1, =□ ・・・・・・ (4) R
.

determined by v5 is the voltage regulator 29° resistor R
2, Rs, R, -------R,, switch S,, S
4...Determined by Sl. In other words, the process is performed based on V input to the microcomputer 19.

is determined, and in this process, switch S, , S4...
・When ■ becomes large, such as 1 or more created by turning on some Sl, it is a big work.

When VW becomes small, a small value 11 can be supplied to the sensor.

tjSl　5 shows one circuit configuration showing another embodiment.

The output of the sensor in this embodiment, which may be the same as in FIGS. 5 and 7, is that the temperature of the solid electrolyte 1 is T,>Tf>T,
...>T, as it becomes lower, Fig. 15(a)
The relationship between the air-fuel ratio λ and the output v1.1 changes as shown in FIG. In this relationship, the air-fuel ratio λ is the X axis, the output ■, 1. For example, when approximated by the following equation, x=m, y+n when the temperature is T1.

At temperature T2 x=m, y+n.

When the temperature is T1, x=m and y+n.

When the temperature is T, X ” m t 3/ + n t
A coefficient of 1 or more m4... m, , nL...
...n.

is stored in advance in the microcomputer.

Import the held V into the computer.

As shown in FIG. 15(b), when vl is larger than a certain value v1 when determining the air-fuel ratio λ.

Using the coefficients mt, n, when v,>vm≧V.

By changing the coefficient for determining λ using 5v1 such as using m2 and n, the correct air-fuel ratio λ can be detected no matter what value the temperature T takes.

Note that a multidimensional approximation formula has better accuracy, so it is also possible to use this approximation formula.

FIGS. 16 and 17 show other temperature compensation circuit configurations. In this embodiment, Ih - Iar tb is fixed.

Temperature compensation is performed by changing E and L. That is, when the temperature is low, E aL<E-' and EIL
Temperature compensation is performed by increasing the sensing time and increasing the sensing time. FIG. 16(a) shows the waveform during sensing, and FIG. 6(b) shows the waveform during biasing. 1st
As shown in Figure 6 (a), one lap is 9I! The temperature is T, l
In this case, let bEaL be E,,'.

Sensing time becomes longer.

In FIG. 17, the OP amplifier 30 generates V.

E to add to. The value of is changed by the output of the microcomputer 19. The held vl is taken into the microcomputer 19. El determined by this V is produced by the switch S1°S 0 . . . Sl operated by the output of the micro combi coater 19. By changing the EEL with V in this way, an output that is not affected by temperature can be obtained.

FIGS. 18 and 19 show temperature function signals instead of vl during sensing (FIG. 18(a)).

The final terminal voltage v during biasing (Fig. 18(b)),
This is another example using . This vs. too.

As can be seen from equation (1), it is proportional to rI.

This is because P(4b) is in P(4a).

Figure 19 shows ■1. This is an example of a circuit that holds . During bias operation, Q of the multivibrator 12 is ON. Therefore, the switch 31 is also turned on, and the capacitor 32 is constantly charged with the terminal voltage V during biasing. Since the switch 31 is turned off when the bias operation is completed, the capacitor 32 remains charged with the final value of the terminal voltage. This value is output as vlM via the buffer amplifier 33. This V□ can be used as the temperature function signal of the temperature compensation circuit configuration described above.

FIG. 20 shows an example of a configuration for obtaining a temperature function signal. On top of the solid electrolyte 1, in addition to electrodes 4a and 4b for measuring the air-fuel ratio, there are also electrodes 1! for measuring internal resistance. pole 3
5a and 5b are provided. A constant current source 36 causes I2 to flow between the electrodes 35a and 35b.

In this case, the electrode 35a is used as a positive electrode, oxygen is caused to flow from the atmosphere to the exhaust side, and the internal resistance of the solid electrolyte 1 is determined from one constant current value and a voltage value at this time. Since oxygen flows from the atmosphere side to the exhaust side, internal resistance can be measured even if the oxygen concentration in the exhaust is low. Further, the value of step 2 at this time needs to be smaller than the limit current value due to the diffusion resistance component of the atmospheric passage. This temperature function signal is input to the microcomputer 37 to perform heater control and correction as described above. FIG. 20 shows an embodiment in which correction is added to the bias time.

FIG. 21 is an embodiment showing another configuration for obtaining a temperature function signal. In other words, the average value V of the terminal voltage at the time of measurement
, y is a temperature function signal.

ff! The solid line waveform shown in FIG. 21(a) is the terminal voltage, and the value shown by the dotted line is the averaged value v1w used as the temperature function signal. This average value v,, is.

The microcomputer 37 digitally performs time integration. One embodiment shown in FIG. 21(b) has a simpler configuration, in which only the terminal voltage at the time of measurement is inputted by a switch 38 into an integrating circuit made of a resistor 39 and a capacitor 40, and the waveform is integrated. . Then hold this value.

Microcomputer 3 via buffer amplifier 41
Enter 7. These input values are v, v in FIG. 21(a).
Since the value is almost the same as 1, it can be used as a temperature function signal. Based on this v, , it is used for heater control and correction as described above. The embodiment shown in FIG. 21(b) changes the bias time.

As shown in FIG. 22(a), a third period t0 is provided only for measuring the temperature.

At this time, a constant current is applied to the solid electrolyte 1 during the period t0, and the internal resistance is measured. In FIG. 22(b).

Switch 6a, bii-OFF, switch 7a between 1c
, b is turned on to flow oxygen from the atmosphere into the exhaust gas, and the second t. The terminal switch 42 is turned on, the terminal voltage between te is held by the capacitor 43, and is input to the microcomputer 37 via the buffer amplifier 44. The microcomputer 37 uses this signal for heater control and pre-existing correction.

FIG. 23 shows another configuration for obtaining a temperature function signal, in which an exhaust gas temperature sensor 47 is provided in the exhaust pipe 46 downstream of the engine 45, and the signal from this sensor 47 is used as the temperature function signal. This signal is input to the microcomputer 37 and a correction signal is output to the drive circuit 48 of the air-fuel ratio sensor 49.

FIG. 24 shows another embodiment, in which an intake air amount sensor 50
And, the operating condition of the engine 45 is determined by the intake pipe negative pressure sensor 51 and the one revolution speed sensor 52! ! (rotational speed and load), and using this operating state as a temperature function signal, the microcomputer sends a correction signal to the drive circuit 48 of the air-fuel ratio sensor 49 to perform temperature compensation.

〔Effect of the invention〕

According to the invention, it is not affected by ambient temperature.

Since an output corresponding to the air-fuel ratio can be obtained, there is an advantage that detection accuracy is improved.

[Brief explanation of the drawing]

FIGS. 1 to 6 are diagrams for explaining the basic principles of the present invention, and FIGS. 7 to 24 are diagrams showing various embodiments of the present invention. 1... Solid electrolyte, 2... Diffusion resistor, 3... Heater. 4...'a pole, 9... Hold circuit, 19... Microcomputer. Agent: Patent attorney Katsuo Ogawa. Was 3 Figure (C) Ball Figure 4 Otsutsu s ('b) Rich χ-1 Thorn $55i1 Su 6 Mark also Wa Mouth 8 Field C (2) (A)
(Ninfu (b)
(c Nori・7-+1・U
Lean λ ￥79 Figure (α〕 (1:)) (
(:) 10th mouth 11th figure 1z figure 130 (α 15th day λO 16th 0λ 17th figure 18th
(shi)Q>Bond #20 23rd edition

Claims (14)

[Claims] 1. an oxygen ion conductive solid electrolyte, first and second electrodes provided on both sides of this solid electrolyte, and this first
In a sudden fuel ratio detector equipped with a diffusion resistor provided on an electrode and exposed to a measurement gas, oxygen is sent from the second electrode side to the first electrode side via the solid electrolyte, and then the first Oxygen is extracted from the electrode side via the solid electrolyte to the second electrode side, and the oxygen concentration in the measurement gas is measured by an output signal indicating the movement of oxygen corresponding to the oxygen concentration in the measurement gas at the time of oxygen extraction. means for outputting a signal that is a function of the temperature of the solid electrolyte; and means for compensating for the effect of ambient temperature on the solid electrolyte in response to the temperature function signal from the output means. Air fuel ratio detector. 2. The air-fuel ratio detector according to claim 1, wherein the temperature function signal is obtained as an output signal during oxygen withdrawal. 3. 2. The air-fuel ratio detector according to claim 1, wherein a heater is provided on the second electrode side, and the compensating means controls the current supplied to the heater. 4. In claim 1, the compensating means comprises:
Air-fuel ratio detection characterized in that temperature compensation is performed by terminating the supply of oxygen when the voltage value when the oxygen is supplied to the first electrode side reaches the voltage value obtained from the temperature function signal. vessel. 5. In claim 1, the compensating means comprises:
An air-fuel ratio detector characterized in that temperature compensation is performed by changing the period during which oxygen is sent to the first electrode side using the temperature function signal. 6. In claim 1, the compensating means comprises:
An air-fuel ratio detector characterized in that temperature compensation is performed by changing a current value applied to the solid electrolyte in order to send oxygen to the first electrode side based on the temperature function signal. 7. In claim 1, the compensating means comprises:
The present invention is characterized in that the relationship between the air-fuel ratio and the detector output at each temperature of the solid electrolyte is stored in advance, and the air-fuel ratio is determined based on the stored relationship based on the temperature function signal to obtain an output that is free from temperature effects. Air-fuel ratio detector. 8. In claim 1, the compensating means comprises:
An air-fuel ratio detector characterized in that the variation width of the terminal voltage that determines the oxygen extraction time width is made variable by the temperature function signal to perform temperature compensation. 9. 2. The air-fuel ratio detector according to claim 1, wherein the temperature function signal is obtained from an output signal of a solid electrolyte when oxygen is sent to the first electrode. 10. 2. The air-fuel ratio detector according to claim 1, wherein the temperature function signal is obtained by measuring an internal resistance provided on the solid electrolyte. 11. An air-fuel ratio detector according to claim 1, characterized in that the temporal average value of the terminal voltage is used as the temperature function signal. 12. The air-fuel ratio detector according to claim 1, characterized in that a third period is provided for measuring the internal resistance of the solid electrolyte, and the internal resistance value measured during this period is used as a temperature function signal. . 13. An air-fuel ratio detector according to claim 1, characterized in that the air-fuel ratio detector measures the exhaust gas temperature of an engine to obtain a temperature function signal. 14. An air-fuel ratio detector according to claim 1, characterized in that the temperature function signal is obtained from the engine speed and load.