JPS6138454B2 - - Google Patents

Description

[Detailed description of the invention]

[Industrial Application Field] This invention relates to a camera exposure control device, particularly
This invention relates to an exposure control device for a TTL direct metering camera. [Prior Art] The TTL direct metering method of a camera converts reflected light from a film or shutter curtain into a photocurrent using a light-receiving element, integrates this photocurrent, and calculates the amount of exposure. In such a photometry method, since the light reflected from the shutter curtain and the film is measured while the shutter curtain is running, accurate photometry cannot be performed due to the difference in reflectance between the two. Therefore, conventionally, a method has been used in which the reflectance of the shutter curtain and the film surface are set to be the same and photometry is performed. [Problems to be Solved by the Invention] However, since the reflectance of the film surface differs depending on the type, it is difficult to perform highly accurate exposure control when using a film that has a different reflectance from the shutter curtain. I couldn't do it. Japanese Unexamined Patent Publication 1973-
Publication No. 46725 solves this problem, but since the first and second photometric circuits are provided to solve this problem, the circuit configuration becomes complicated and the level adjustment of the two photometric systems is required. There is a problem that requires The present invention was made in view of the above problems, and an object of the present invention is to provide an exposure control device for a camera that can automatically correct the difference in reflectance between the shutter curtain and the film using a single photometric circuit system. do. [Means for Solving the Problems] According to the present invention, as shown in FIG. 1, a photoelectric conversion element measures reflected light from at least one of the shutter front curtain surface and the film surface and outputs a photocurrent. 101, a photocurrent amplifying means 102 for amplifying the photocurrent output from the photoelectric conversion element by a predetermined amplification factor and outputting a photocurrent substitute characteristic value corresponding to the amplified photocurrent; A/D conversion means that A/D converts the photocurrent substitute characteristic value at least once before exposing the film surface, and repeatedly A/D converts the photocurrent substitute characteristic value immediately after exposing the film surface. 108, storage means 103 for storing an A/D conversion output value obtained from the A/D conversion means before the film exposure as an initial value (V _B2S ), and an A/D conversion output of the A/D conversion means. value (V _B2 ) and the initial value (V _B2
S ) immediately after the film surface is exposed, and the A/D conversion output value (V _B2 ) and the initial value (V _B2
a comparing means 104 that generates an output when the magnitude relationship with _S ) is reversed; and receiving the output of the comparing means, the amplification factor of the photocurrent amplifying means is set to a predetermined value so that the amplified photocurrent is kept approximately constant. an amplification factor correcting means 105 for inhibiting a decrease in the amplification factor immediately after the front shutter curtain is fully opened; and an integrating means 106 for integrating the amplified photocurrent of the photocurrent amplification means.
and a shutter control means 107 for starting the running of the shutter trailing curtain based on the output of the integrating means. [Operation] The present invention will be explained with reference to FIGS. 2A, 2B, and 3. When measuring light incident on a camera, the incident light is received by a light receiving element, such as a silicon photodiode SPD, as reflected light from one or both of the shutter front curtain and the film.
The photocurrent from the SPD is amplified by a photocurrent amplification circuit, then integrated by an integration circuit, and output as a photometric signal. By the way, while the front shutter curtain is running, it is necessary that the integrated current is always constant, but the amount of light received by the SPD changes as shown in FIG. 2A depending on the position of the front shutter curtain. That is,
Photometry is started, and at time t0, the amount of light received from the front of the shutter front curtain is shown, at time t1, the amount of light received from the front shutter curtain and the film is shown, and at time t2, the amount of light received from the front of the film is shown. In this way, when the photocurrent corresponding to the amount of received light that changes depending on the position of the shutter front curtain is amplified and integrated, a constant integrated current cannot be obtained. Therefore, when the photocurrent amplification factor of the photocurrent amplification circuit is changed as shown in FIG. 2B, the integrated current becomes constant as shown in FIG. 3. [Embodiment] The camera exposure control device shown in FIG. 4 is composed of a photocurrent amplification circuit 1, a constant current circuit 2, an integration circuit 3, and an amplification factor control circuit 4. In the photocurrent amplifier circuit 1, the anode and cathode of a silicon photodiode SPD are connected to the inverting and non-inverting terminals of an operational amplifier _A1 , respectively. The non-inverting terminal of operational amplifier _A1 is grounded and the inverting terminal is connected to the collector of transistor _Q1 . The emitter of this transistor _Q1 is connected to the output terminal of the operational amplifier _A1 via a resistor _R1 , and is also connected to the emitter of a transistor _Q2 . The base of transistor _Q1 is grounded. The collector and base of transistor _Q2 are connected to the non-inverting input terminal of operational amplifier _A2 . The inverting input terminal of operational amplifier _A2 is connected to the collector and base of transistor _Q3 , and the output terminal is connected to resistor _R2.
is connected to the emitters of transistors Q ₃ and Q ₄ through. The base of the transistor Q ₄ is grounded, and the collector is connected to the inverting input terminal of the operational amplifier A ₃ of the integrating circuit 3. A parallel circuit including a capacitor C and a field effect transistor _Q5 is connected between the inverting input terminal and the output terminal. Op amp A ₃
The non-inverting input terminal of is grounded, and the output terminal is connected to the shutter control circuit 6 via the determination circuit 5. The current amplification factor control circuit 4 is provided with a microcomputer 11, a DA converter 12, and a DA converter 13. D-A converter 12
The output terminal of is connected to the non-inverting input terminal of operational amplifier _A4 of constant current circuit 2. The inverting input terminal of this operational amplifier _A4 is connected to Vcc via a resistor _R3 , and the output terminal is connected to the gate of a field effect transistor _Q7 . The source of this transistor _Q7 is connected to Vcc via a resistor _R3 , and the drain is connected to the base of transistor _Q6 . The collector of this transistor Q ₆ is connected to the resistor R ₃ , and the emitter is connected to the collector and base of the transistor Q ₂ of the photocurrent amplification circuit 1 . Further, a resistor _R4 is connected between the base and emitter of the transistor _Q6 . A-D converter 13 of the amplification factor control circuit 4
is connected to the base of the transistor Q ₂ of the photocurrent amplifier circuit 1 and the non-inverting input terminal of the operational amplifier A ₂ . The gate of transistor Q ₅ connected to capacitor C of integrating circuit 3 is connected to switch driver 7, which is controlled by microcomputer 11. In the exposure control device shown in FIG. 4, the photocurrent Ii of the silicon photodiode SPD is amplified according to the ratio of the first constant current I _REF and the second constant current I' _REF as shown in the following equation. That is, I′i=(I′ _REF /I _REF )Ii (1) Here, the first constant current, that is, the output current I of the constant current circuit 2
_REF is calculated as follows. I _REF = (Vcc-Va)/R ₃ (2) However, when Va is substituted for the output voltage of the D-A converter (2) into the equation (1), it becomes as follows. I'i=R ₃ I' _REF / (Vcc・Va)・Ii...
...(3) Since the second constant current I' _REF and the voltage Va are independent variables, from equation (3), by changing the output voltage Va of the D-A converter connected to the computer, the photocurrent amplification factor, that is, I' _REF /I Understand that _REF can be easily changed. Here, the voltage V _B2 is expressed by the following equation. V _E2 =kT/qln( _IREF /Ii) =kT/qln(Vcc-Va/ _IiR3 )...(4
) Therefore, the voltage V _B2 and the photovoltage Ii, and the voltage V _B2 and the photocurrent amplification factor information voltage Va all correspond to each other in a 1:1 ratio, and since the voltage V _B2 is a value obtained by logarithmically compressing the photocurrent Ii, it can be applied over a wide range. Highly accurate photometry is possible for photocurrent levels of . Next, the photometry method of the present invention will be explained with reference to the flowchart of FIG. When the camera release button is pressed, the mirror begins to rise. When the mirror lift is completed, the photocurrent amplification factor information (Va) stored in advance is output. Next, the substitute characteristic value V _B2 of the photocurrent is sampled from the photocurrent amplification factor control circuit 4. In this sampling, since the optical response speed of the silicon photodiode SPD is fast, if only one sampling is performed, an error will occur due to fluctuations of fluorescent lights, etc., so multiple samplings are performed. When the photocurrent substitute characteristic value V _B2 is sampled n times, the average value V _B2S of the n samplings is calculated. When the calculation is completed, a front curtain start signal is output. When time t3 has elapsed since the output of the front curtain start signal, the average value V _B2S calculated earlier is greater than the predetermined voltage V _B2 t, that is, the brightness of the subject becomes dark.
When the photocurrent of the SPD is small, the amplification factor of the photocurrent amplification circuit is controlled based on photocurrent amplification factor information stored in advance. By doing so, leakage current of the SPD and errors in performance of the photocurrent amplifier circuit are prevented from increasing. The average value V _B2S is the specified voltage V _B
2 When the value is smaller than t, switch driver 7
turns off the analog switch, i.e. transistor _Q5 .
and start the integration. Here, a delay time is provided between the start of the leading curtain and the start of integration, which compensates for errors caused by the response delay from the output of the leading curtain run start signal until it actually starts running, and the inertia of the curtain. It is for the purpose of Next, the substitute characteristic value V _B2 is sampled at a certain period determined by the calculation speed of the D-A converter, so that it is always equal to the previously calculated average value V _B2S until a certain second time t ₄ when the curtain is fully opened. The photocurrent amplification factor information is sequentially calculated and output. In normal photometry in which the light source of the subject does not change during shooting, the photocurrent Ii gradually increases as the camera shifts to film photometry. On the other hand, the substitute characteristic value V _B2 gradually becomes smaller as the current Ii becomes larger, as shown by the above-mentioned equation (4). Therefore, in the actual calculation, if the substitute characteristic value V _B2 is smaller than the average value V _B2S calculated initially, it is first confirmed that Va<Va ₁ , and then the photocurrent amplification information Va is incremented and output. If the average value V _B2S is V _B for the substitute characteristic value V _B2
If the relationship is _2S < V _B2 , the previous photocurrent amplification factor information
Va is output as is. Here, the photocurrent amplification factor information Va is compared with Va ₁ at the end of the calculation of the memory formula t.
This is to reduce the error in the TTL direct photometry section due to abnormal light that enters just before time = _t4 . This Va ₁ is determined by the reflectance of the curtain and film, and it must be able to correct the photocurrent amplification factor for all films so that it can be used as a memorized photometer without error until the second when the curtain is fully opened. Therefore, Va ₁ must be selected at a value such that the above objective is achieved when the curtain has the smallest reflectance and the film has the largest reflectance. In addition, in the above, O is calculated as a memory formula.
In the interval <t<t ₄ , if the subject brightness becomes darker than the initial value (at the time of sampling the initial value of the substitute characteristic value), the photocurrent amplification factor information Va does not change and follows the change in subject brightness. It is similar to photometry and does not cause any errors in exposure. The control of the photocurrent amplification factor as explained above is the fourth
This is carried out by the photocurrent amplification factor control circuit 4 shown in the figure, and this control circuit is constructed as shown in FIG. Next, how the photocurrent amplification factor is controlled will be explained with reference to this circuit. When the mirror is released, the mirror starts to rise, and when the mirror completes its rise, the A port of PIA (Peripheral Interface Adapter, which will be described later) is activated by the mechanical switch shown in Figure 7.
The terminal of PA ₇ is changed from L level to H level. At this time, when the computer detects the mirror rising completion, the initial photocurrent amplification factor information Va stored in the ROM is
Output. This information is led to the DA converter 12 via the B port of the PIA. This information is D
- When converted into analog information by the A converter 12, the substitute characteristic value V _B2 is sampled n times.
It is stored at the specified address in RAM. Here D
- There is no need to check the conversion end flag of the A converter. This is because the conversion speed of the D-A converter is so high that the conversion is completed within the execution of one instruction by the MPU (microprocessor unit). Next, the average value V _B2S is calculated by multiplying the sum of the n substitute characteristic values V _B2 by 1/n and written at the designated address. Next, PIA
The first curtain starts when the output terminal PB6 of the B port of is changed from the L level to the H level. If the time since the start of the first curtain has not reached the specified time _t3 , the program will further delay the time, and when t= _t3 , the previously calculated average value V _B2S will be changed to a certain value stored in the ROM. If it is larger than V _B2 t, that is, if the subject brightness is very dark and the error is large,
The initial value of the photocurrent amplification factor stored in the ROM is output, and then a gate signal is applied to the analog switch _Q5 shown in FIG. 4 to turn off the switch _Q5 and start integration. This is the output terminal of PIA's B port
This is done by changing PB7 from L level to H level. In order to ensure the switching operation of analog switch _Q5 , the signal of PB7 is applied to the gate of analog switch _Q5 via switch driver 7. The circuit configuration of this switch driver 7 is shown in FIG. In this diagram
When PB7 is at H level, the emitter of transistor _Q9 is drawn into negative power _VEE , and analog switch _Q5 is
The signal is turned OFF, and the integral capacitor C becomes capable of integration. When PB7 is at L level, the emitter of transistor _Q9 becomes equal to power supply voltage Vcc, analog switch _Q5 is turned on, and no charge is accumulated in capacitor C. Here, the low resistance R ₅ is a bias resistance between the gate and the source, and the transistor Q ₉ is a reverse voltage blocking transistor. When integration is started as described above, ROM photocurrent amplification factor information is output at certain time intervals. The delay time t ₃ from the start of the first curtain to the start of integration in the above is to compensate for errors caused by the mechanical response delay of the magnet and the inertia of the curtain, and this delay time t ₃ is automatically set during the manufacturing process. wanted
It is stored in ROM to save labor and improve accuracy. If it is determined that the average value V _B2S is smaller than the predetermined value V _B2 t and no error will occur even if the photocurrent amplification factor is controlled based on the information on the subject brightness, integration is started in the same way as above, and the computer A substitute characteristic value V _B2 determined by the _photocurrent amplification factor information _Va outputted in The value is compared with the value Va ₁ stored in the ROM, and if Va > Va ₁ , Va
is decremented, and if Va<Va ₁ , the previously output amplification factor information Va is output. This operation
This is repeated until a certain time _t4 stored in the ROM, that is, until the curtain is fully opened. The photocurrent amplification information Va is guided to the D-A converter via the B port of the PIA. The number of pits of the D-A converter is A-
It depends on the conversion processing speed of the D converter and the program execution speed of the computer. Also, this D-
The number of bits of the A converter also affects the exposure accuracy of the camera, so it depends on the conversion processing speed of the D-A converter, the program execution speed of the computer, and the D-A converter.
The number of bits of the A converter must be selected with these considerations in mind. The faster these execution speeds are, the better the exposure accuracy will be. In the example, A-D
The processing speed of the converter is 100 μsec, and the D-A converter has an 8-bit configuration. In this case, the amplification factor information Va can be changed by 10 steps in 1 msec, and sufficient exposure accuracy can be obtained. Assuming that the number of steps for the photocurrent amplification factor Va to change from the value of Va at the start of the first curtain to the lower limit value Va ₁ determined by the film reflectance is 14, it takes 1 msec.
The number of steps that can be changed during this time is at most 10 steps, so if there is a change in the amount of light equivalent to more than 10 steps within 1 msec, the calculation will not be able to keep up and an error will occur. Therefore, the number of steps required for Va to change from the value of Va at the start of the first curtain to the lower limit value _Va1 must be determined from the assumed brightness distribution of the subject. If the brightness distribution is extremely uneven, the photocurrent will change drastically in a short period of time when transitioning from curtain metering to film metering.
If the number of Va steps is too large, the above error may occur. For example, there is a spot-like object that emits strong light with a radius of 3 mm on the image plane.
If the surrounding area is completely black, the photocurrent will change from around the initial value of Va to the lower limit value Va ₁ in about 1 msec, and the total number of steps of Va and the processing per unit time of the A-D converter will change. The larger the ratio of times, the larger the error. However, the above-mentioned subject is rare, and even with such a subject, if the processing speed of the A-D converter is set to 100 μsec and the number of steps of Va is set to 14, it will not cause a fatal exposure error, and it will be possible to photograph a normal subject. This will give you sufficient accuracy. In the above explanation, we considered the case of a subject with abnormal brightness, but if the brightness distribution of the subject is uniform and average photometry is used, we can use a mathematical formula to consider how the photocurrent Ui changes. The photocurrent amplification factor stored in is calculated assuming that the luminance distribution of the subject is uniform. If the horizontal length of the camera screen is l, and the reflectance ratio between the curtain and the film is α, then when the front curtain travels a distance x after reaching the screen, the photocurrent is as follows compared to full-curtain metering. is multiplied by a value like . xα+(l−x)/l (5) Therefore, in order to make the photocurrent amplification factor equal to the initial photocurrent Ii, it must be multiplied by the following value. A ₁ =l/+xα+(l-x) ......(6) Figure 9 shows the change in photocurrent amplification factor when α=8 and l=36mm, and the change in photocurrent amplification factor is calculated using A as the vertical axis and x as the horizontal axis. It becomes like this. Since x is a time function, on the horizontal axis,
Also shown is the time t since the first curtain appears on the screen.
On the other hand, if Va is set to 14 steps and the initial value of (Vcc-Va) is 10 mV and each step is 5 mV, Vcc-Va and photocurrent amplification factor A ₂ (from equation (3), A ₂ = R ₃ I' _REF /Va) The relationship with is shown in Figure 10. When creating this graph, assume that the ratio of reflectance between curtain and film is 1:8 at maximum, and when (Vcc - Va) = 80 mV, the photocurrent amplification factor is set to 1/8 of the initial value. . From FIG. 9, when the subject brightness is uniform and average photometry is used, the maximum value of the change in photocurrent amplification factor during 1 msec is 0.4 times the initial value. On the other hand, as shown in Fig. 10, when the photocurrent amplification factor information Va changes by two steps from 10 mV to 20 mV, the photocurrent amplification factor changes by about 0.5 times, and since the changing speed of the A-D converter is 100 μsec, these two steps It can be seen that the conversion requires only 200 μsec, which is a processing speed that can sufficiently follow changes in photocurrent. Furthermore, since the ratio of reflectance between the curtain and the film is large, the change in the magnitude of the photocurrent is greatest when the front curtain begins to appear on the screen, and gradually decreases, as can be seen from FIG. On the other hand, when the photocurrent amplification factor information Va is varied arithmetic one step at a time, the change in the photocurrent amplification factor is largest at first and gradually becomes smaller. By doing this, the change in photocurrent amplification factor can be tracked even when there is a large change in photocurrent, and the integral current I′i can be kept constant with a very small error, resulting in exposure accuracy that has almost no practical problems. It will be done. Figure 11 shows the microprocessor unit used in the photocurrent amplification factor control circuit of Figure 6.
The circuit configuration of the MPU is shown, and an outline of this MPU will be explained below. This MPU is provided with many internal registers, and the values of these registers are set by instructions. The program counter has the role of storing the address to be executed in order to execute the program in order. This program counter increases one by one from the smallest memory address to the largest memory address in the order of execution.Accumulators A and B temporarily store data to shorten the instruction program. All data in the registers is processed using this accumulator.
For example, when executing the instruction x-Y=S, if an accumulator is not used, the address storing X, Y, and S will require 6 bytes, and the + operation will require 1 byte. Code required. In this example, if X is placed in the accumulator and S is placed in the accumulator, the machine code that must be included as an instruction will only need to be Y, thus requiring only 3 bytes in total. The condition code register is a register that stores the state of the operation result. The computer determines the state of this register and executes the instruction. This includes flags for performing conditional branches, such as whether a carry or borrow has occurred as a result of CPU interrupt function control and calculations, whether the result is zero, or whether an overflow has occurred. . The ALU is the part that executes instructions such as clearing, repeating memory contents (COM), addition, and subtraction. Accumulators are frequently used. There are also many cases where you want to use an accumulator without destroying its contents during a program. In such a case, the original value is saved to some memory, and when the use of the accumulator is finished, the original value that was saved earlier is returned from memory. In this case, a save address is given to each register, and the data is stored in the memory at the address indicated by this value. The stack pointer is a register that indicates the address to store. The instruction register is a register for storing the state of external control terminals and controls the operation of the MPU. Each control input terminal has the following function. Reset: Resets or starts the MPU at power-off or initial startup. NMI: Abbreviation for Non-Maskabie Interrupt, and when this input changes from H level to L level, it causes the MPU to perform an NMI operation. Interrupts from this pin are CCR
Accepted regardless of the contents of the interrupt flag (condition code register). Holt: This is a Revesence input and is normally kept at H level to keep the MPU in GO state. When the level reaches L, the MPU stops operating at the end of the currently executing instruction.
The bus available becomes "H", the valid memory address VAA becomes "L", and the address bus (A ₀ to A ₁₅ ), data bus (D ₀ to D ₇ ), and read/write are all in a three-state state. Bus available: Under normal operating conditions, it is “L” (BA). When the halt signal is "L" and the device is in the halt state, or when the wait command is executed and the device is in the wait state, this BA terminal becomes "H". The "H" output of the bus available signal indicates that the MPU is in a stopped state and that the address bus is available for other uses. Data bus enable: Used for data bus three-state (DBE) control, and when set to "L", the data bus drive is in operation. Valid Memory Address: This output indicates to the memory or (VMA) peripheral that there is a valid signal on the address bus. In normal operation, the AND signal of VMA and _φ2 is used as an enable signal for peripheral interfaces such as memory or PIA. Read/Write: Indicates whether the MPU is in a read or write state for memory or verification. Normally, this signal is at the "H" level (read state) and becomes the "L" level only when writing data to the memory or peripheral. The above is an explanation about MPU, but next is PIA
This will be explained with reference to FIGS. 12 and 13. Interfaces PIA MPU and peripheral devices. This PIA has four control lines and two 8-bit bidirectional data bus lines (A and B buses) between the peripherals and can interface with the MPU and most peripherals. Interfacing is possible. The operational functionality on the PIA is determined by programming the contents of the PIA's control registers from the MPU. As shown in FIG. 12, this PIA has two symmetrical blocks, each of which is provided with an interrupt controller. The two sets are called A ports and B ports, and both can be used as input ports and output ports. However, when performing handshake control, the A port is limited to input and the port is limited to output. Hereinafter, when the A port and B port are commonly explained, the symbol X will be used. There are eight registers inside the PIA, and the process by which the CPU accesses these registers is normally done by a signal on the signal line RS ₁ connected to the output buffer A ₁ of the MPU, as shown in Figure 13. The A and B ports are selected and the signal line is normally connected to A ₀ .
RS ₀ selects the control register CRX and data-related registers. Then, data direction register DDRX or peripheral data register is selected based on the value loaded into CRX, which will be described next, and finally an input/output port is selected by R/W. This R/
The input/output port selection by W can be performed regardless of the input/output switching of the RXn terminal, which will be described below. PIA
The peripheral device side pins are set as input/output pins by the value set in DDRX. This is the value of each bit 7 to 0 of DDRX, and the “0” bit is
Input/output switching is performed one by one, such that PXn becomes an input and PXn with a "1" bit becomes an output. Control registers CRA and CRB are roughly divided into a part that represents status and a part that controls PIA. Bit 2 of CRX is “0”
When PAn or PBn of the data-related registers
Select DDRX to determine the direction of. CRX ₂ =
When it is "1", an input port or an output port is selected. Therefore, the PIA setting is CRX ₂ “0”
PXn by loading the value into DDRX with
The direction of the input/output port is determined and then the input/output port is called by setting this bit to "1".
CRX 0 and 1 control the CX ₁ terminal.
The CX ₁ input is used for flag input during data transfer using flags; when it is an input port, it is used as a data ready flag from a peripheral device; when it is an output port, it is used as a data accepted flag from a peripheral device. This input is auto-reset by reading the input/output port as described below, so it is a dynamic input and the flip-flop inside the PIA is set by the edge of this input.
Output to CRX ₇ . CRX ₁ is a control bit that determines the edge on which this flip-flop is set, and is set to "0" for a negative edge and "1" for a positive edge. This flip-flop is always output to CRX ₇ and a flag check is performed by reading this. CRX ₀ is a bit that allows interrupt requests to the CPU by setting a flag flip-flop. The input/output of the CX ₂ pin is determined by CRX ₅ , and when that bit is “0”, it is input.
It can be used in the same way as the CX ₁ input terminal. In this case, the flag from the CX ₂ terminal is output to CRX ₆ . The controls for this are CRX ₄ and
This is done by CRX ₃ and is the same as the control of CX ₁ terminal. CX ₂ terminal by programming 1
When used as a bit output port, CRX ₅ =
“1”, CRX ₄ = “1” is set. In this way, the CX ₂ terminal outputs the data written to CRX ₃ and can be used to output flags by the program. The MPU and PIA have been described above, but what happens when a photocurrent amplification control circuit including such an MPU and PIA is actually operated according to a program will be described below. In this example, Motorola's
It uses the instruction format of the M6800 microprocessor. When the reset terminal of PIA is set to "L", all bits below the input port become "0". As a result, all the terminals of the A port and B port of the PIA are set as input ports, so it is necessary to set the terminals again. First, all of PA ₀ to PA ₇ of the A port of the PIA are input ports, and the CA ₁ input has a positive edge active and interrupts are disabled, and the CA ₂ outputs "H" constantly.
The CA ₁ input of the PIA is connected to the EOC terminal (End of Conversion) of the A-D converter. This EOC
The terminal changes from "L" to "H" to notify that the conversion has been completed. The 7th bit CRA ₇ of the control register at port A of PIA is set to "1" as CA ₁ changes from "L" to "H". The conversion speed of the A-D converter is
Since it takes 100 μsec, this CRA ₇ flag is checked to see if it is OK to input the V _B2 information. In order to perform the above functions, the A port control register of PIA has LDAA and STAA.
The instruction loads 3E (hexadecimal). MPU
Since the calculation speed is faster than the conversion speed of the A-D converter, there is no particular need to check whether reading of the data register of the PIA is completed.
Since the B port of PIA is set as an output port, all data direction contents are set to "1".
At this time, the contents of the control register are the same as those of the A port. LDAA, STAA, etc. indicate mnemonics of instruction codes. These instruction codes are shown in the programs shown in Table 1, and the operands and address formats used in these programs are shown in Tables 2 and 3. It is shown. All operation codes are in hexadecimal, binary, decimal, and
The hexadecimal correspondence is shown in Table 4. However, in the case of hexadecimal, the mark S is added at the beginning. LDAA is an instruction to temporarily load S3E into the MPU's accumulator.
Address S00B1 of the port control register is called and S3E is stored. This means that the initial setting of the A port has been completed. Next, the contents of the data direction register of the B port are all overwritten and set to "1" by the COM instruction, thereby setting the B port as an output port. Next, the contents of the control register of the B port are initialized in the same manner as in the case of the A port.
In the above, the flowchart shown in Figure 14
PIA initial settings are completed. Subroutines are often used in programs to reduce the amount of memory, but before jumping to a subroutine in the middle of a program, it is necessary to memorize the address to which the program will return when returning from the subroutine. The memory address setting for storing this address is performed using the LDS instruction. In this embodiment, it is set to address S00E3. When the initial setting program is completed in this manner, a photometry operation is performed next. in this case,
When the mirror rises, the A port's PA7 terminal changes from "L" to "H". This is TST and
Judgment is made by BPL command. The TST instruction is an instruction that subtracts S00 from the contents of the register.
The BPL instruction means a command to return to the loop 1 branch again if the most significant bit corresponding to the PA7 terminal is "0". When the mirror rises and the most significant bit of the data register becomes "1", the loop is exited and the next LDAA instruction is executed. Here, the initial photocurrent amplification factor information Va stored at address S00B5 (hereinafter the upper byte S00 of the address will be omitted) is loaded into the accumulator. Next, the contents of the accumulator are PIA
The photocurrent amplification factor is controlled via the D-A converter and the constant current circuit. The above is the program up to address S15. Next, in LDX and CLR shown in the flowchart, the number of samplings of V _B2 is set and the contents of the V _B2 storage memory are cleared. In this case, the sampling period is approximately 1 msec, and sampling is performed eight times. To do this, first S0008 is loaded into the index register, and then the contents of address SE0, which stores the results of each sampling, are cleared. When the conversion is completed, the A-D converter is connected to the CA1 terminal, and the bit of CRA7 of the control register is changed from "0" to "1" at the active edge from "L" to "H" through EOC. The instructions and judgment for this purpose are at address S1D.
This is being done at TST and BPL. TST instruction is PIA
This is an instruction to subtract S00 from the contents of the control register of , and in the next BPL instruction, whether the contents are positive or negative is determined by displaying the contents in two's complement. Specifically, the most significant bit of the control register
The TST and BPL programs are repeatedly executed until CRA7 becomes "1".
When the CRA7 bit becomes “1”, the address of the data-related register of the A-D converter is changed through PIA.
SB0 is called and A-D converter output information V
_B2 is loaded into the accumulator. this is
This is done using the LDAA instruction at address S22. Next, the contents of the accumulator at address SE0 are added and stored at address SE0 again. In addition, the information for the 8th time is
They may be stored at different addresses in RAM and then added, but in this case the number of memories increases. Next, the contents of the index register are decremented by the DEX instruction at address S26. If this result is not 0 by the BNE command, that is, if eight samplings have not been completed, the JMP command is used to create a delay time corresponding to the sampling period. First, the routine in this case will be explained. When the result of the BNE instruction at address S27 is not 0, the routine moves to address S2C using the JMP instruction. S2E~
The program at address S34 stores delay time information at addresses SE1 and SE2. In the example
S0078 is stored. JSR at address S36 is an instruction to move to a subroutine, and a delay time of approximately 1 msec is obtained from this subroutine. This time calculation will be explained in detail in the last subroutine program. After the predetermined time has elapsed, the subroutine returns and the JMP command at address S39 causes loop 2 to repeat the sampling eight times. If it is confirmed that Z=0 by the BNE instruction at address S27, move to the JMP2 branch V _B2
The average value is calculated from the information added eight times. This can be easily done by shifting each digit of the addition information three digits to the right. LSRA is an instruction to execute this, and this instruction is from S3D address to S3E
By performing this three times at the address, the contents of the accumulator are shifted three places to the right and the average value is determined. Next, by the STAA instruction at address S3F
The average value V _B2S of V _B2 is stored at address SE0 in the RAM. When the average value V _B2S is determined, the first curtain is started, and this is done by changing the output terminal PB6 of the B port of the PIA from "L" to "H". For this purpose, the LDAA instruction (address S41)
S40 (0100 0000 in binary), which is the data for setting PB6 to "H", is loaded into the accumulator, and S40 is stored in the output register of the B port of PIA by the STAA instruction. For the reasons mentioned above, a delay time of several ms is required from the start of the first curtain until the exposure time is actually counted. Note that the start of counting the exposure time corresponds to the start of integration. To create a delay time, first the delay time information is loaded into the accumulator using the LDAA instruction, and then stored at addresses SE1 and SE2 using the STAA instruction. When the first curtain starts, a delay time is created to move to the subroutine, but in this subroutine,
Information on addresses SE1 and SE2 can be obtained. However, this delay time varies depending on the camera, and is corrected by counting manual seconds.
This delay time can be automatically calculated by linking the shutter tester and the camera. Therefore, this delay time information is calculated during the assembly process and stored at a specially provided address in a nonvolatile memory, PROM, or the like. This allows automation. In order to actually realize this, special measures are required, but since this invention is not intended to achieve this, average delay time information will be stored. The JSR instruction at address S4D branches to a subroutine for creating the above-mentioned delay time, and after a predetermined time elapses, the process returns to address S50. Since the subject is very dark and the observation error must not be large, the previously determined average value V _B2S is compared with the upper limit value V _B2 t, and if V B2S is small, the upper limit value V _{B2 t} is determined in advance according to the flowchart shown in FIG. When V _B2S is large, the photocurrent amplification factor is controlled according to the flowchart shown in FIG. 15 so that a constant integrated current is obtained. To do this, first the contents of address SE0, where V _B2S is stored, are loaded into the accumulator by the LDAA instruction at address S50, and then by the CMPA instruction.
V _B2 t stored in the ROM is compared and a decision branch is made at the BGT instruction. If ZU(NV)=
0, that is, if V _B2S > V _B2 t, the process moves to address S59 and is controlled by a predetermined photocurrent amplification factor. If ZU (NV) = 1, it branches to address S7A and automatically increases the photocurrent amplification factor. controlled. In the program from address S59, LDAA,
The STAA instruction outputs the initial photocurrent amplification factor information stored at address SB5 in the ROM. At the same time, integration starts, but this is the PIA
This is done by changing the PB7 terminal of the B port from "L" to "H". At this time, the photocurrent amplification factor information must not be changed, so information obtained by adding data S80 for setting the PB7 terminal to "H" is stored at address SB5 and output. Similarly to the SB5 address, a value obtained by adding S80 to the photocurrent amplification factor information is stored at addresses SB7 to D5. The B port of the PIA also controls the front curtain start as described above, so the D-A converter is controlled by the lower 6 bits. 1 by program
Although the front curtain start and integration start signals can be controlled with bits, four bits are sufficient for controlling the D-A converter, so this is done. When the integration starts, the photocurrent amplification factor information of the ROM is sequentially output from address SB7 to address SD5 at a cycle of about 1 msec. For this reason, first, the leading address S00B7 of the photocurrent amplification factor information is loaded into the index register. This is done by the LDX instruction at address S5D. Next, the LDAA instruction is executed, but since this instruction is in index address format,
The operand of the LDAA instruction is the address (index register value + offset value S00). Therefore, initially, the contents of the first address SB7 of the photocurrent amplification factor information are loaded into the accumulator. At address S6F, the contents of the index register are incremented by INX, so when this command is executed again, the photocurrent amplification factor information at address SB8 is output for the second time. Photocurrent amplification factor information is output by the STAA command at address S62. S64 to S6A
As mentioned above, the address determines the sampling period when moving to the subroutine, and here S0078 is stored at addresses SE1 and SE2 to obtain a sampling period of approximately 1 msec. S6C
After branching to a subroutine by the address JSR
When the return is made to address S6F, the value of the index register loaded at address S5D is incremented, and preparations are made for calling the address where the next photocurrent amplification factor information is stored. The value of the index register and S00D6 are compared by the CPX instruction, and if they are equal, the flag Z of the condition code register of the MPU becomes 1, and the program moves to the BRA instruction at address S78 and stops. Z＝
If it is 0, it branches to address S60, and the photocurrent amplification factor information in the ROM is called and controlled every sampling period. The above describes the control program when the subject brightness is dark and the error is considered to be large. Next, a program will be described in which it is determined that the brightness of the subject is set to a certain value to be brighter and no exposure error will occur. If ZU (NV) = 1 at address S54, by the LDAA instruction that branches to address S7A
The value at address SB7 is loaded into the accumulator. As mentioned above, this address SB7 stores photocurrent amplification factor information and integration start information, and the following
Integration is started by outputting the STAA command to address SB2 of the B port of PIA, and the photocurrent amplification factor signal is output. In this routine program, the photocurrent amplification factor is automatically controlled, but it cannot follow rapidly changing light such as strobe light, so it is set at a certain time when the curtain is fully open (time t ₄ in Figure 5). ) it is necessary to fix the photocurrent amplification factor. As an initial setting for counting this time, the value of S012C is loaded into the index register by address S7E. This value is decremented each time the conversion end signal of the A-D converter is output by the DEX command at address S98.The value of the index register becomes S0000.The photocurrent amplification factor is fixed at the value output last. . Since the A-D converter used has a conversion speed of approximately 100 μsec, approximately 30 msec is obtained by performing S12C (300 in decimal) times. A-
To determine whether the D conversion is completed, S00 is subtracted from the contents of the PIA B port by the TST command at address S81, and the most significant bit of the result is set to "0" by the next BPL command.
If so, the program branches to the TST instruction at address #81 again, and if the most significant bit is "1", the program moves to address S86. At this address S86, the content V _B2 of the data related register of the A port of PIA is loaded into the accumulator, and in the next instruction, that value is compared with the value stored at address SE0 and judged by the BLT instruction at address S8A. A branch is made. If NV=1, that is, V _B2S > V _B2 , branch to address S8F and PIAB
The contents Va of the port data related registers are loaded and stored at address SB6 by the next CMPA instruction.
It is compared with the upper limit value Va ₁ of Va, and if ZU (NV) = 1, that is, Va > Va ₁ , by the BLE instruction at address S93.
The process branches to address S96 and the value of Va is output unchanged. ZU(NV)=0 or Va>Va ₁
If so, move to address S95, the value of the accumulator is decremented, and Va is decreased by one step.
The data is stored in the data-related register of the B port of PIA by the STAA instruction. Next, the process moves to address S98, and the contents of the index register are decremented.
If Z=1, that is, the content of the index register is "0000", the value of Va is fixed by the BEQ instruction. If Z=0, the process branches to address S7A again. Next, to explain the subroutine, the STX instruction at address SA1 is an instruction to save the contents of the index register in order to use it within the subroutine without destroying the contents of the index register before moving to the subroutine. . The time information stored at addresses SE1 and SE2 is loaded into the index register by the LDX instruction. SA5 address
The value of the index register is decremented by the DEX instruction, and a decision branch is made by the next BNE instruction. If the result of the BNE instruction is not Z=0, that is, the content of the index register is "0000", the program branches to address SA5 again. If Z=1
The value of the index register saved at address SA1 is restored by the LDX instruction, and returned to the main program by the RTS instruction at address SAA. here
When S0078 is stored in addresses SE1 and SE2, the amount of delay time obtained by the subroutine instruction is calculated. MPU clock frequency is 1μ
sec, the execution time of each instruction is as follows. STX: 5μsec LDX: 4μsec DEX: 4μsec BNE: 4μsec RTS: 5μsec Therefore, LDX and DEX commands are executed S0078 times (in decimal
120 times), 5+4+(4+4)×120+4+5=978 μsec is obtained. [Effects of the Invention] As described above, in the present invention, a photocurrent substitute value before the leading start of the leading curtain is stored, and based on this value, the integral current output from the photocurrent amplification means is set to the initial value. The amplification factor is controlled so that they are equal. This initial value does not depend on the difference in light reflected from the film only from the front shutter curtain, and therefore automatically corrects the reflectance of the film. Moreover, 1
Since the outputs of the two optical conversion elements are used for storage before the front curtain starts running and for real-time measurement after the start of running, the circuit configuration is simplified and the troublesome adjustment is eliminated.

【table】

【table】

【table】

【table】

【table】

【table】

【table】

【table】 [Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of the present invention, and FIGS. 2A, 2B, and 3 show the amount of received light, photocurrent amplification factor, and Integral current curve diagram, 4th
The figure is a circuit diagram of an exposure control device for a camera according to the present invention.
Fig. 5 is a flowchart of the camera exposure control device, Fig. 6 is a circuit diagram of the photocurrent amplification factor control circuit of the camera exposure control device of Fig. 4, Fig. 7 is a switch circuit diagram, and Fig. 8 is a switch circuit diagram. The circuit diagram of the driver, FIG. 9 is a graph showing the relationship between the current amplification factor and the shutter curtain traveling position, FIG. 10 is a graph showing the multiple of the photocurrent amplification factor with respect to the photocurrent amplification factor information, and FIG. The MPU circuit diagram, Figure 12, is
PIA circuit configuration diagram, Figure 13 is a flowchart explaining the operation of PIA, and Figures 14 to 16 are
The figure is a flowchart showing the operation process of the camera exposure control device according to the present invention according to a program. 101...Photoelectric conversion element, 102...Photocurrent amplification means, 103...Storage means, 104...Comparison means, 105...Amplification factor correction means, 106...Integration means, 107...Shutter control means, 108... …
A/D conversion means.

Claims (1)

[Claims] 1. A photoelectric conversion element that measures reflected light from at least one of the shutter front curtain surface and the film surface and outputs a photocurrent, and a photoelectric conversion element that amplifies the photocurrent output from the photoelectric conversion element by a predetermined amplification factor. a photocurrent amplifying means for outputting a photocurrent substitute characteristic value corresponding to the photocurrent; A/D converting means for repeatedly A/D converting the photocurrent substitute characteristic value immediately after exposing the film surface, and an A/D conversion output value obtained from the A/D converting means before exposing the film to an initial value. (V _B2S ), the A/D conversion output value (V _B2 ) of the A/D conversion means and the initial value (V _B2S ) are compared immediately after the film surface is exposed, and the Comparing means for generating an output when the magnitude relationship between the D-converted output value (V _B2 ) and the initial value (V _B2S ) is reversed; amplification factor correction means for reducing the amplification factor of the photocurrent amplification means by a predetermined value and prohibiting the reduction of the amplification factor immediately after the shutter front curtain is fully opened; and integrating the amplified photocurrent of the photocurrent amplification means. 1. An exposure control device for a camera, comprising: an integrating means for calculating the output of the integrating means; and a shutter controlling means for starting the movement of a rear shutter curtain based on the output of the integrating means.