CS7622 CCD Imager Analog Processor Features Description l 13-Bit The CS7622 is a low-power analog front-end processor for interline or frame transfer CCD imagers. Main applications include digital still image cameras and video cameras. A/D Conversion Using DRX™ Technology l Backlight Compensation l Supports Full Scale Analog Input Voltage Ranges from 300 mV to 1 V in 100 mV Increments l High Resolution Output Mode l Low Resolution (Preview) Output Mode for LCD Driver l Integrated Correlated Double Sampler l Digital Black Level Clamp l Digital Outputs Selectable for 13, 12, or 10 Bits l Low Power Consumption l Power Down Mode l High Speed Serial Interface l Supports a Large Variety of Clock Input Frequencies l Low power mode option CCD OUTPUT CDS/DRX GAIN The architecture includes a correlated double sampler, black level clamp and a 13-bit A/D conversion module using patented DRX technology. Chip parameters can be programmed using a high speed 4-wire asynchronous digital interface. The chip outputs digitized CCD data in either 13-bit, 12bit or 10-bit format. 10-bit outputs are generated from the 13-bit A/D output by a programmable companding curve. ORDERING INFORMATION CS7622-IQ -40 to +85 °C 32-pin TQFP 7x7x1.4m A/D CONVERTER OUTPUT COMPANDER CLOCK BLACK LEVEL DATA OUT CLOCK OUT REGISTER BLOCK CK_FT CLOCK GENERATOR CK_DATA SERIAL INTERFACE SERIAL BUS Preliminary Product Information P.O. Box 17847, Austin, Texas 78760 (512) 445 7222 FAX: (512) 445 7581 http://www.cirrus.com This document contains information for a new product. Cirrus Logic reserves the right to modify this product without notice. Copyright Cirrus Logic, Inc. 1999 (All Rights Reserved) JUL ‘99 DS322PP1 1 CS7622 TABLE OF CONTENTS 1.0 CHARACTERISTICS/SPECIFICATIONS ........................................................... 4 1.1 DIGITAL CHARACTERISTICS.................................................................... 4 1.2 POWER CONSUMPTION ........................................................................... 4 1.3 RECOMMENDED OPERATING CHARACTERISTICS............................... 4 1.4 ABSOLUTE MAXIMUM RATINGS .............................................................. 4 1.5 ADC (ANALOG-TO-DIGITAL CONVERTER).............................................. 5 1.6 CDS/VGA PARAMETERS........................................................................... 5 1.7 SERIAL INTERFACE TIMING SPECIFICATIONS ...................................... 5 2.0 GENERAL DESCRIPTION .................................................................................. 7 3.0 OPERATION ........................................................................................................ 8 3.1 CDS/VGA (correlated double sampling/variable gain amplification) ........... 8 3.2 Black Level Adjustment ............................................................................ 10 3.3 Gain Adjust Block ..................................................................................... 11 3.4 13-to-10 Bit Compander ........................................................................... 12 3.5 Stand By and Preview Mode ................................................................... 14 3.6 Serial Interface .......................................................................................... 14 3.7 Input Timing for Sampling Clocks ............................................................. 15 4.0 REGISTER DESCRIPTIONS ............................................................................. 17 Reset ........................................................................................................ 18 Power down Control 1 .............................................................................. 18 Operation Control 1 .................................................................................. 19 Operation Control 2 .................................................................................. 20 Black Level Control (8 LSBs) .................................................................... 20 Black Level Control (MSB) ........................................................................ 21 Black Level Control - General ................................................................... 21 Black Level Control - Loop Gain, Clamp Length ....................................... 22 Gain Calibration Offset 1 .......................................................................... 23 Gain Calibration Offset 2 .......................................................................... 23 Gain Calibration Offset 3 .......................................................................... 23 Fixed Gain ................................................................................................ 25 Compander - Black slope, Slopes (MSBs) ............................................... 27 Compander Slope 1 (LSBs) ...................................................................... 27 Compander Slope 2 (LSBs) ...................................................................... 27 Compander Slope 3 (LSBs) ...................................................................... 28 Compander Slope 4 (LSBs) ...................................................................... 28 Compander Offset 1 ................................................................................. 28 Compander Offset 2 (MSBs) .................................................................... 29 Compander Offset 2 (LSBs) ..................................................................... 29 Compander Offset 3 (LSBs) ..................................................................... 29 Compander Offset 4 (LSBs) ..................................................................... 29 Contacting Cirrus Logic Support For a complete listing of Direct Sales, Distributor, and Sales Representative contacts, visit the Cirrus Logic web site at: http://www.cirrus.com/corporate/contacts/ Preliminary product information describes products which are in production, but for which full characterization data is not yet available. Advance product information describes products which are in development and subject to development changes. Cirrus Logic, Inc. has made best efforts to ensure that the information contained in this document is accurate and reliable. However, the information is subject to change without notice and is provided “AS IS” without warranty of any kind (express or implied). No responsibility is assumed by Cirrus Logic, Inc. for the use of this information, nor for infringements of patents or other rights of third parties. This document is the property of Cirrus Logic, Inc. and implies no license under patents, copyrights, trademarks, or trade secrets. No part of this publication may be copied, reproduced, stored in a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photographic, or otherwise) without the prior written consent of Cirrus Logic, Inc. Items from any Cirrus Logic website or disk may be printed for use by the user. However, no part of the printout or electronic files may be copied, reproduced, stored in a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photographic, or otherwise) without the prior written consent of Cirrus Logic, Inc.Furthermore, no part of this publication may be used as a basis for manufacture or sale of any items without the prior written consent of Cirrus Logic, Inc. The names of products of Cirrus Logic, Inc. or other vendors and suppliers appearing in this document may be trademarks or service marks of their respective owners which may be registered in some jurisdictions. A list of Cirrus Logic, Inc. trademarks and service marks can be found at http://www.cirrus.com. 2 DS322PP1 CS7622 Compander X1 (MSBs) ............................................................................ 30 Compander X1 (LSBs) ............................................................................. 30 Compander X2 (MSBs) ............................................................................ 30 Compander X2 (LSBs) ............................................................................. 30 Compander X3 (MSBs) ............................................................................ 31 Compander X3 (LSBs) ............................................................................. 31 Device ID .................................................................................................. 31 Revision Code .......................................................................................... 31 5.0 PIN DESCRIPTIONS ......................................................................................... 32 Supply ...................................................................................................... 32 Ground ..................................................................................................... 32 CMOS Input ............................................................................................. 32 CMOS Analog Input ................................................................................. 33 CMOS 4 mA Output ................................................................................. 33 6.0 PACKAGE DIMENSIONS ................................................................................. 34 LIST OF FIGURES Figure 1. SEN Timing.........................................................................................................................6 Figure 2. Serial Write Timing..............................................................................................................6 Figure 3. Read Data Timing ...............................................................................................................6 Figure 4. Digital Camera Block Diagram............................................................................................7 Figure 5. CS7622 Block Diagram.......................................................................................................7 Figure 6. Idealized CCD output waveform .........................................................................................8 Figure 7. Transfer function of VGA circuit (assuming full scale level of 1.0 V) ..................................9 Figure 8. Block diagram of CDS/VGA circuit......................................................................................9 Figure 9. Idealized timing diagram of VGA/CDS circuit ...................................................................10 Figure 10.Black level adjustment loop ..............................................................................................11 Figure 11.Transfer function of Vin to Gain Adjust output Block (assuming full scale level of 1.0 V).12 Figure 12.Gain Adjust output Block...................................................................................................12 Figure 13.13-to-10 bit compander.....................................................................................................13 Figure 14.CS7622 output data and clocks ........................................................................................14 Figure 15.Input Timing ......................................................................................................................14 Figure 16.Typical Connection Diagram.............................................................................................16 Figure 17.Transfer Function of Analog Input to Digital Output (assuming full scale level of 1.0 V) ..24 Figure 18.Transfer Function of ADC with Fixed Gain Settings (assuming full scale level of 1.0 V)..26 DS322PP1 3 CS7622 1.0 CHARACTERISTICS/SPECIFICATIONS DIGITAL CHARACTERISTICS (TA = 25 °C; VDDD = 3.3 V) Parameter Logic Inputs High-Level Input Voltage Low-Level Input Voltage Input Leakage Current Logic Outputs High-Level Output Source Current @ IOH = 4 mA Low-Level Output Sink Current @ IOL = 4 mA 3-State Leakage Current Symbol Min Typ Max Units VIH VDD-0.8 - - V VIL - - 0.8 V IIN - - 10 mA VOH VDD-0.4 - VOL - - 0.4 mV IOZ - - 10 µA mV POWER CONSUMPTION (TA = 25 ° C; VDDA = VDDD = 3.3 V; Output Load = 30 pF; Input Clock = 15MHz) Parameter Symbol Min Typ Max Units Power Dissipation Peak Mode Preview Mode Stand By Down PD PDLR PDPD - 214 162 0.0825 - mW mW mW Analog Power Supply Current Peak Mode Preview Mode Stand By Down IAN IALR IAPD - 53 37 0.025 - mA mA mA Peak/Preview Mode Preview Mode IDN IDPD - 12 0 - mA mA Symbol Min Typ Max Units VDDA VDDD 3.0 2.5 3.3 3.6 3.6 V V 10 mV Digital Power Supply Current RECOMMENDED OPERATING CHARACTERISTICS Parameter Power Supply Voltage GNDA to GNDD Voltage Differential Analog Full Scale Input Voltage Range AIN 300 mV - 1V Vp-p - 20 MHz - MHz Input Clock Rate ABSOLUTE MAXIMUM RATINGS Parameter Power Supply Voltage Symbol Min Max Units VDDA, VDDD -0.3 6.0 V GNDD-0.3 VDDD+0.3 V GNDA-0.3 VDDA+0.3 V 10 mA -0 +70 °C +260 °C +150 °C Digital Input Voltage Analog Input Voltage Input Current AIN (except supply pins) Ambient Temperature Range Lead Solder Temperature (10sec duration) Storage Temperature Range -65 WARNING: Operation at or beyond these limits may result in permanent damage to the device. Normal operation is not guaranteed at these extremes. 4 DS322PP1 CS7622 ADC (ANALOG-TO-DIGITAL CONVERTER) Parameter Symbol Full Scale Input Voltage Range Min Typ 300 mV Full Scale Input Voltage Range Resolution Max Unit 1V V0-p 100 mV ADC resolution - 10 - bits Total Differential Non-Linearity - ±1 - LSB Total Integral Non-Linearity - ±1 - LSB Min Typ Max Unit 1V V0-p CDS/VGA PARAMETERS Parameter Symbol Input Voltage Range 300 mV Total Gain Range Input Referred Noise (rms) Maximum Gain Setting AVGA - 18 - dB VnVGA - - 0.2 mV SERIAL INTERFACE TIMING SPECIFICATIONS Description Enable Setup SDAT Setup SDAT Hold Serial Clock Period Write Data Invalid Read Data Valid Clock to Disable SEN Rise to SEN Fall (Note 1) Symbol t1 t2 t3 t4 t5 t6 t7 t8 Minimum 10 10 10 143 0 0 143 200 Maximum 10 10 - Unit ns ns ns ns ns ns ns ns Notes: 1. the minimum serial clock period must be longer than two pixel clock periods. DS322PP1 5 CS7622 SEN t1 t7 t8 SCLK SDATI R/W, ADDR <6.0> DATA <7.0> Figure 1. SEN Timing SEN t4 SCLK SDATI R/W t2 A6 A5 A6 A3 t3 Figure 2. Serial Write Timing SCLK t5 SDATI SDATO t6 XX (DON’T CARE) A0 D7 D6 D5 Figure 3. Read Data Timing 6 DS322PP1 CS7622 2.0 GENERAL DESCRIPTION The CS7622 forms the heart of a four chip digital CCD Camera. The four chips include the CCD imager, the CS7622 CCD digitizer, a vertical drive interface chip and a backend DSP chip to further process the digital data (see Figure 4.) The patented DRX technology allows the CS7622 to output data with 13-bit dynamic range, and at the same time reducing the power consumption to a 10bit equivalent A/D converter. The digitized output is either available in 13-bits, 12-bits or 10-bits. The 10-bit output is created by companding the 13-bit A/D output to 10-bits. The companding curve consists of 4 linear segments, where each slope and each start point is user programmable. A block diagram of the CS7622 chip is shown in Figure 5. CS7622 CDS/ADC Video Output Backend DSP CCD Control LCD Panel Vertical Drive Timing Signals +5 V +5 V to -5 V DC-DC converter Figure 4. Digital Camera Block Diagram VDD[2] GND[2] AIN Σ CDS/VGA A/D Gain Adjust 13 to 10-bit Compander M U X Black Level CK_FT CK_DATA Clock Generator Reference Serial Interface SEN DOUT[12:0] (up to 3 may be unused) CLKO CLAMP TEST RST REF_CAPP 1 µF REF_CAPN BG_RES 10 kΩ SDATI SDATO SCLK Figure 5. CS7622 Block Diagram DS322PP1 7 CS7622 PIXEL PERIOD VIDEO SIGNAL RANGE RESET LEVEL FEED THROUGH LEVEL DARK VIDEO LEVEL MAX. BRIGHTNESS Figure 6. Idealized CCD output waveform 3.0 OPERATION 3.1 CDS/VGA (correlated double sampling/variable gain amplification) An idealized waveform of the CCD output is shown in Figure 6. The CCD output contains reset noise, thermal noise, and 1/f noise generated in the CCD output circuit. This degrades the S/N ratio and must be cancelled. Since the noise during the active video portion of the CCD signal is assumed to be correlated with the noise during the feed through portion of the signal, this noise can be cancelled by subtracting the feed through level from the video level. This operation is called correlated double sampling. The active video signal is the difference between the feed through and video levels. The active video signal varies according to light conditions. In order to insure that the full dynamic range of the ADC is utilized even under low light conditions, the CCD output is amplified using a VGA. The gain control is provided by a 2 bit control word generated by an ADC after stage 1, which has a gain of 1. Based on the input voltage, a gain of 1x, 2x, 4x, or 8x is subsequently applied to the signal. The amount of gain is later adjusted in the digital section. After the VGA, the signal gets digitized by a 10 bit ADC. The 2 bit ADC output is used in 8 combination with the 10 bit ADC output to produce a 13 bit output. Adding more gain before the ADC does not offer performance improvement because the noise of the CCD (after gain is applied to it) begins to dominate over the quantization noise. Any additional gain should be done in digital since the performance is the same as when the ADC output has the additional gain applied. In order to add more flexibility, the full scale input range is programmable through register 05h. This setting will determine what input level maps to the highest ADC output code. Thus depending on the saturation level of the particular CCD used in the system, an appropriate full scale input level can be chosen in the CS7622. The choices of full scale input level are 300 mV to 1 V in 100 mV increments. In the remainder of this document, all the figures and discussions assume a full scale level of 1 V is used. The transfer function of the VGA portion of the circuit is shown in Figure 7 with full scale level = 1 V. It is assumed that the CDS has already been performed. If desired, the gain switching functionality can be disabled and forced to a fixed gain of 8x, 4x, 2x, or 1x. This way any dynamic range enhancement is lost and the digital output is only 10 bits. If DS322PP1 CS7622 VOUT (V) 1.07 0.5 8X 0.125 00 4X 2X 0.5 0.25 01 1X 1.0 10 11 VIN (V) ADC OUTPUT Figure 7. Transfer function of VGA circuit (assuming full scale level of 1.0 V) Φ2 Φ1 C3 C1 100 KΩ Φ1 C1 C5 C2 C4 VOUT VIN -A1 100 KΩ Cb VREF -A2 Vo1 Vo2 STAGE 1 STAGE 2 ADC 2 -A3 STAGE 3 CONTROLS C3, C5 CONTROLS GAIN ADJUST BLOCK IN DIGITAL Figure 8. Block diagram of CDS/VGA circuit a fixed gain of 1x is selected, DOUT[12:3] is used as the output, a fixed gain of 2x will use DOUT[11:2], etc. In order to use this mode, the fixed gain register (14h) should be set and the calibration offset registers (OEh - 10h) should be set to 0. The CDS/VGA circuit is composed of three stages. The first stage has a fixed gain of 1, and the second and third stages have variable gain with a combined gain range of 1 to 8 (0-18 dB). Figure 8 shows a block diagram of the CDS/VGA circuit. The total gain is A = (C2/C3)(C4/C5) which is adjusted by varying C3 and C5. The capacitor Cb on the front DS322PP1 of stage 1 is for black level adjustment and will be discussed in detail later. This circuit utilizes a two phase non-overlapping clock to perform the desired CDS function. The two phase clock also allows the video signal to be passed to the output while retaining a positive polarity signal. Figure 9 shows a timing diagram of the two phase clock along with the CCD signal and output signals of stages one, two and three. There is an internal mid-scale DC bias level circuit at the input pin. This allows AC coupling into the CS7622 with a capacitor and having the input auto- 9 CS7622 matically biased to mid-supply without worrying about external circuitry to perform this task. 3.2 Black Level Adjustment In order to maintain a constant reference level for black pixels, a feedback loop is implemented that sets the black level value at the output of the ADC to 64 in the 13 bit digital code. This loop is active during the optically black pixels which are output at the beginning and end of a frame as well as during a portion of the horizontal blanking period. The presence of black pixels in the CCD output is indicated by the CLAMP pulse, which is supplied externally through the CLAMP pin. The black level can also be written to through the serial port. In order to acquire a starting value for the black level, the loop will run over the several lines of black pixels at the beginning of the frame. The block diagram of the loop is shown in Figure 10. The update rate is once per line during active pixel lines as long as the Clamp pulse is < n+10 cycles. Where n is the number of pixels accumulated before the black loop is updated and is programmable through register 0Dh bits 5:0. If the Clamp pulse is longer CCD INPUT SIGNAL than n+10 cycles the black loop is updated every n+10 cycles. For example, during optical back lines the loop is updated several times at a rate of once every n+10 cycles. The open-loop transfer function of the black level adjustment loop is K×n H ( z ) = ------------z–1 blk_gain = 1, 2, 4, or 8 where blk_gain is programmable through a register and n = # of black pixels during clamp time, which is also programmable. The value of Kxn will determine the open-loop gain of the system. The settling time for the loop can be calculated using the following formula: For offset range=1 (reg 06h, bit 0) 1 1 τ = – --------------------------- ----- ln ( 1 – nK ) fu For offset range =0 1 1 τ = – ---------------------------- ----- fu ln 1 – nK ------- 2 V(2) V(1) 1 K = --------- blk_gain 256 V(3) ck_ft ck_data OUT OF STAGE 1 OUT OF STAGE 2 OUT OF STAGE 3 V(1) V(2) V(1) V(3) V(2) V(1) V(3) V(2) Figure 9. Idealized timing diagram of VGA/CDS circuit 10 DS322PP1 CS7622 During fixed gain mode the time constant is a little different. 1 1 τ = – ---------------------------- ----- fu ln 1 – nK ------- 8 For a fixed gain of 2: 1 1 τ = – ---------------------------- ----- fu ln 1 – nK ------- 4 For a fixed gain of 4: 1 1 τ = – --------------------------- ----- fu ln 1 – nK ------- 2 For a fixed gain of 8: 1 1 τ = – -------------------------- ----- ln ( 1 – nK ) fu 3.3 Gain Adjust Block In order to increase the dynamic range of the ADC, a variable gain, whose value is determined by the signal level, is applied to each pixel. This allows for 13 bits of dynamic range and 10 bits of resolution after accounting for the significance of the ADC output bits. The gain applied in the analog is illustrated in the transfer curve in Figure 7. Once the signal is digitized, the gain adjust block uses the gain information for a given pixel word and shifts its bits accordingly. For example, using a full scale level of 1.0 V, if Vin = 0.3 V, the VGA would choose a gain of 2X so the ADC input is 0.6 V. The 10-bit output of the ADC (with no black level) is (0.6/1.0) × 1024 = 614, or “1001100110.” in binary. The gain adjust block will take this value plus the bits representing the 2x gain and divide the output by two (shift right by 1). The output of the gain adjust block is then “0100110011.000.” Note that the decimal point is virtual, having no existence in silicon. It is representing the fact that we keep 3 extra bits of lower significance in the output. In the same manner, if Vin = 0.75 V, a gain of 1X would be chosen and the output of the gain adjust block In order to achieve no ringing in the settling use, n n ---- ≤ 1 for offset range = 1, and ------≤ 1 for offset range K 2K = 0. The 9 MSBs of the black level accumulator can be read or written through a register. If written, the LSBs are set to zero. The black level is set to “8” in a 10-bit digital output representation. In a 13-bit representation, it is set to “64.” The power-up default value in the accumulator is at mid level. ‘64’ VIN 10 Σ CDS/VGA ADC - + FROM SERIAL INTERFACE BLK LVL LOOP GAIN REG CLIP 7 K + FU = UPDATE FREQUENCY FP = PIXEL FREQUENCY Z-1 Z-1 FP FU DAC + MUX For a fixed gain of 1: Also note that the black level adjust loop can be disabled. In addition, the black level can be programmed through the serial port. BINARY TO THERM 9 Figure 10. Black level adjustment loop DS322PP1 11 CS7622 would be “1100000000000.” The transfer function of the Vin/gain adjust out is shown in Figure 11. A block diagram of the gain adjust block is shown in Figure 12. Since the analog gain changes do not match the digital shifts exactly, there is a potential to have non-monotonic digital output. In order to remove this problem, calibration is performed. During calibration, offset values are found that will be used to counteract the errors caused by the analog gain mismatch. Using these offset values, the final output is a monotonic continuous 13-bit value. 3.4 13-to-10 Bit Compander While a 13 bit output may be useful in some applications, others may require the standard 10 bit output. To accommodate this and yet still retain the advantages of the increased dynamic range, a 13to-10 (or 13-to-12) bit compander is included. By using the picture content as a guide, the user can select which curve will lead to the best overall dynamic range in the picture. The Companding module takes 13-bit data as input, and outputs either 10-bit companded data, 12-bit MSB-clipped data or it lets the original 13-bit data pass through. By programming the compander in the way that is shown in Figure 13, it is possible to compensate for DIG ADJUST OUT (13 BITS) 8192 4096 8X 4X 2X 1X 2048 1024 0 0.25 0.125 00 01 0.5 1.0 10 11 VIN (V) ADC OUTPUT Figure 11. Transfer function of Vin to Gain Adjust output Block (assuming full scale level of 1.0 V) ADC OUTPUT 10 VGA_ADC OUTPUT 2 GAIN ADJUST 13 TO DIGITAL GAIN SHIFT BY 0,1,2, OR 3 Figure 12. Gain Adjust output Block 12 DS322PP1 CS7622 backlighting conditions. Details in dark areas stay visible, even in very complex lighting conditions. These three modes can be selected through 2 register bits in operational control. Bits_out register bits 0x 10 11 Output mode 10 bits companded 13 bits 12 bits (clipped) ble offset value, offset1. This may be set to 0 if desired. This option will lose the “blacker-thanblack” pixel information, but allow for slightly more dynamic range. Note: If using the linear mode (option 1), offset1 must be set to 8. Registers x1 through x3 should be programmed with the x coordinates of each one of the three knees. Table 1. In the 12-bit clipped mode, any input above 4095 gets clipped to 4095. In the 10-bit companded mode, the input gets companded through a four segment, three knees, fully programmable curve. Registers slope1 through slope4 should be programmed with 256 multiplied by the calculated slopes. Finally, the offsets can be programmed following the formulas below: To program the curve, the placement of the three knees in the companding curve must be determined. The next step is to determine the slope of the four segments created by the three knees (slope for each segment is defined as delta y / delta x). Finally, offsets must be calculated to keep the companding curve continuous. y1 = slope1/256 × (x1-64) + offset1 A fourth knee exists in the curve, which represents the black level value. There are two options for the 10-bit black value. In case one, a linear mapping is employed such that “blacker-than-black” pixel information is kept, with black (code 64 in the 13 bit data) being defined as code 8 in the 10 bit domain. The second option clips all pixel values less than black (code 64 in the 13 bit data) to a programma- offset4 = y3 - (x3 × slope4 / 256) y2 = slope2/256 × (x2-x1) + y1 y3 = slope3/256 × (x3-x2) + y2 offset2 = y1 - (x1 × slope2 / 256) offset3 = y2 - (x2 × slope3 / 256) (use integer division and discard the remainder) When using the 10 bit companded output, be aware of the non-linearity of the output data. If linear output is needed to perform Auto White Balance (AWB) or Automatic Gain Control (AGC), a linear curve can be implemented to gather statistics. This can be achieved by writing 8191 to x1 (set register CODE_OUT 1023 (x3,y3) OFFSET4 SLOPE4 (x2,y2) SLOPE3 OFFSET3 (x1,y1) SLOPE2 OFFSET2 OFFSET1 SLOPE1 64 X1 X2 X3 8191 CODE_IN Figure 13. 13-to-10 bit compander DS322PP1 13 CS7622 1Fh to 1fh and set register 20h to ffh) and setting slope1 to 32 (set register 15h to 00010xxxb and set register 16h to 20h). Once the statistics have been gathered, all four registers should be returned to their previous values before taking the actual picture. The output of the compander is available at the pins DOUT<9:0> and it makes transitions either at the falling or rising edges of the pixel rate clock CLKO, controlled by a register bit. The Falling edge option is shown in Figure 14. strongly recommend that the chip should be kept in Stand By mode when not in use in order to save power. When in preview mode, a user may wish to cut down the resolution of the ADC output to 6 bits in order to reduce the power consumption of the CS7622. In this mode, the current is reduced by 20 mA. With the DRX (Dynamic Range eXtension) circuitry, 3 bits of dynamic range are added to the 6-bit ADC output producing a 9-bit output. The pins DOUT[12:4] are used to output the digitized data in preview or Stand By mode. 3.5 Stand By and Preview Mode 3.6 Serial Interface In order to enter power down mode a value of 07h must be written to register 01h. This will power down all the analog sections. Stopping the input clocks will power down the digital. To power up again, the input clocks must be turned on first then a value of 00h needs to be written to register 01h. The user must wait at least 500µs for the internal analog references to settle to their appropriate values before normal operation is resumed. It is The serial interface is designed to allow high speed input to control the chip’s registers. The specifications on this interface are as follows: Asserting the enable pin, SEN, enables the serial interface to perform data transfers. Data present on the SDATI pin is latched into the CS7622 on each rising edge of the serial clock, SCLK. Data output on SDATO from the CS7622 is clocked out on the rising edge of SCLK. CLKO DOUT<9:0> Figure 14. CS7622 output data and clocks T1 T4 CCD INPUT SIGNAL CK_FT T3 T2 CK_DT Figure 15. Input Timing 14 DS322PP1 CS7622 The CS7622 receives only the first 16 rising edges of the SCLK while SEN is low and then ignores any remaining SCLK and SDATI information. If SEN goes high before 16 SCLK pulses have been received, the CS7622 aborts the serial transfer. The first bit is the R/W bit. R/W = 1 identifies the transfer as a read. If (0), the transfer is a write. The next seven bits define the address. For write transfers, the second byte of the 16-bit packet contains the data byte. For read transfers, the CS7622 outputs the read data on SDATO after accepting the address. Address and data are transferred MSB first. When not reading out data, the SDATO pin is not driven by the chip (Hi-Z state). The timing diagrams and specifications are shown in “Serial Interface Timing Specifications” on page 5 and Figures 1, 2, and 3 on page 5. 3.7 Input Timing for Sampling Clocks The input clocks CK_FT and CK_DT are used to set up the sampling times and also to generate the internal digital clock. These clocks need to be running when processing pixels from the CCD, writing to the chip registers, or performing calibration (See DS322PP1 Register Description of Operation Control 2 reg 05h bit 0 for the details of performaing a calibration). The timing of these clocks is important to ensure optimum settling times and sampling the correct value. CK_FT and CK_DT need to be nonoverlapping pulses made as wide as possible to give long settling times. The falling edge of CK_FT should be close to the end of feedthrough while the falling edge of CK_DT should be close to the end of the data section of the CCD signal. See figure 15. Typical timing is given in table 2. Timing Parameter Typical Operating Values T1, T4 2 ns T2, T3 5 ns Table 2. Longer non-overlapping values for T1 and T4 will increase the recovery time, thus requiring a slower clock rate. 15 CS7622 VCC 13 28 VAA Sampling Signals CK_FT CK_DT 19 CK_FT 20 CK_DATA 13 DOUT[0:12] CLKO 21 to Mic CS7622 17 RESET NC 16 9 RST DIAG TEST REF_CAPP REF_CAPN 18 from Microcontroller BG_RES CLAMP 15 1 µF 14 10 10 kΩ ±1% 5 SCLK 7 SDATI 6 SDATO 8 SEN from CCD 11 AIN GND 12 29 Figure 16. Typical Connection Diagram 16 DS322PP1 CS7622 4.0 REGISTER DESCRIPTIONS Register (hex) 00h 01h 02h - 03h 04h 05h 06h-0Ah 0Bh 0Ch 0Dh 0Eh 0Fh 10h 11h - 13h 14h 15h 16h 17h 18h 19h 1Ah 1Bh 1Ch 1Dh 1Eh 1Fh 20h 21h 22h 23h 24h 25h 26h Register Function Software Reset Power Down Control 1 Reserved Operation Control 1 Operation Control 2 Reserved Black Level Control - Accumulator (LSB) Black Level Control - Accumulator (MSB) Black Level Control - Loop Gain, Clamp Length Gain Calibration - Offset 1 Gain Calibration - Offset 2 Gain Calibration - Offset 3 Reserved Gain Calibration - Fixed Gains Compander - Black slope, Slopes (MSBs) Compander - Slope1 (LSBs) Compande - Slope2 (LSBs) Compander - Slope3 (LSBs) Compander - Slope4 (LSBs) Compander - Offset1 Compander - Offsets (MSBs) Compander - Offset2 (LSBs) Compander - Offset3 (LSBs) Compander - Offset4 (LSBs) Compander - X1 (MSBs) Compander - X1 (LSBs) Compander - X2 (MSBs) Compander - X2 (LSBs) Compander - X3 (MSBs) Compander - X3 (LSBs) Device ID Rev Code Access W R/W R/W R/W R/W Default value (hex) 00h 00h 00h 0Ah 04h R/W R/W R/W R/W R/W R/W 00h 01h 2Ah 00h 00h 00h R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R/W R R 00h 10h B2h 60h 20h 07h 08h 0Bh BFh 05h 20h 03h 20h 05h 18h 0Bh 58h CCh 00h Table 3. Register Description DS322PP1 17 CS7622 Reset Default = 00h; Read/Write (address 00h) Bit Number Bit Name Default 7 6 5 4 3 2 1 RESERVED 0 Bit Mnemonic 7:1 - 0 sft_rst 0 0 0 0 sft_rst 0 0 0 0 Function reserved Software Reset: When this bit is written with a ‘1’, all of the digital circuitry and the registers will reset to their default values. It automatically clears after 4 pixel clock periods. The clocks remain running during the reset period. Power down Control 1 Default = 00h; Read/Write (address 01h) Bit Number Bit Name Default 18 7 6 5 4 3 RESERVED 0 0 0 0 0 2 1 0 pd_vga pd_adc pd_ref 0 0 0 Bit Mnemonic Function 7:3 - 2 pd_vga DRX Front End Power Down: When written with a ‘1’, the DRX front end circuitry powers down. 1 pd_adc ADC Power Down: When written with a ‘1’, the Analog-to-Digital converter circuitry powers down. 0 pd_ref Voltage Reference Power Down: When written with a ‘1’, the Analog-to-Digital converter circuitry powers down. reserved DS322PP1 CS7622 Operation Control 1 Default = 0Ah; Read/Write (address 04h) Bit Number Bit Name Default 7 Bit Mnemonic 7:6 - 5 dout_edge 4 3:2 1 0 DS322PP1 6 RESERVED 0 5 4 3 2 1 0 dout_edge low_res bits_out1 bits_out0 blk_dis off_range 0 0 1 0 1 0 0 Function reserved This register is used to set when dout changes values. Relative to CLKO 0 - dout output changes on the falling edge of CLKO 1 - dout output changes on the rising edge of CLKO low_res Preview Mode: This mode can be used to cut the current consumption of the chip by 20 mA. The output of the ADC will have 6 bits of resolution in this mode, and the output of the chip will have 9 bits after using the DRX circuitry. It is intended to be used when driving an LCD display or any other time when a lower resolution picture is acceptable. bits_out1-0 Number of Data Bits Out: The range of the output data can be determined by these bits. The data internal to the chip has a 13-bit range. The output can be this full range, half this range (12 bits), or an eighth of this range (10 bits). If 12bit data is selected, the top half of the 13-bit range is saturated to the maximum 12-bit code. If 10-bit data is selected, the compander curve which is user programmable is employed to map the 13-bit data to the 10-bit output. 0 - 10 bits output; 1 - 10 bits output 2 - 13 bits output; 3 - 12 bits output blk_dis Black Level Loop Disabled: If the user chooses to adjust the black level himself through register access, he may disable the internal black level loop. This loop usually updates the black level to what it calculates to be the correct level. If disabled, the offset used will be determined from the value written in the black level accumulator register. 0 - internal black level loop is enabled 1 - black level loop is disabled off_range Offset Range: The black level loop is used to cancel any offsets from the CCD and chip circuitry. If the offsets are small, the user has the option to decrease the offset cancellation range for the added advantage of increasing the resolution of the offset cancellation. 0 - smaller offset cancellation range used (~50 mV) 1 - larger offset cancellation range used (~100 mV) 19 CS7622 Operation Control 2 Default = 04h; Read/Write (address 05h) Bit Number Bit Name Default 7 6 0 0 5 4 RESERVED Bit Mnemonic 7:4 - 3:1 0 0 0 3 2 1 0 fs_lvl2 fs_lvl1 fs_lvl0 gain_cal 0 1 0 0 Function reserved fs_lvl2-0 Full Scale Level: This is used to set the full scale input range of the CS7622. Since CCDs have various saturation levels, it is advantageous to set the full scale input range of the CS7622 to match the saturation level of the CCD used. The table below shows the full scale level choices. (See Table 4) gain_cal Gain Calibration: A calibration of the gain stages is required to insure a monotonic digital output. If the user wishes to initiate a calibration, he may do so by setting this bit to ‘1’, which will invoke a gain calibration sequence immediately. This bit automatically clears itself after a calibration has been initiated. During the calibration sequence the output will not contain valid data. The input clocks must be running throughout the whole calibration sequence which lasts for ~760 clocks. fs_lvl Full Scale Voltage 000 0.3 V 001 0.4 V 010 0.5 V 011 0.6 V 100 0.7 V 101 0.8 V 110 0.9 V 111 1.0 V Table 4. Black Level Control (8 LSBs) Default = 00h; Read/Write (address 0Bh) Bit Number Bit Name Default 20 7 6 5 4 3 2 1 0 accumulator accumulator accumulator accumulator accumulator accumulator accumulator accumulator 7 6 5 4 3 2 1 0 0 0 0 0 0 0 0 0 Bit Mnemonic 7:0 accumulator7-0 Function Black Level Accumulator: See the description of register OCh. DS322PP1 CS7622 Black Level Control (MSB) Default = 01h; Read/Write (address 0Ch) Bit Number Bit Name Default 7 6 5 0 0 0 3 2 1 0 0 0 RESERVED Bit Mnemonic 7:1 - 0 4 0 0 accumulator8 1 Function Reserved accumulator8 Black Level Accumulator: is a 9 bit number representing an amount of offset added to the input of the CDS circuit. The black level loop alters the black level accumulator value to make the output of the ADC settle to code 64 during black pixels. If desired the black loop may be disabled and written to manually to add any desired amount of offset. There is a total of ~100 mV of offset range if the offset range register setting is set to “1” or ~50 mV when this register setting is set to “0”. This offset range is used to correct for CCD offsets plus internal offsets generated in the analog path of this chip. The offset range before subtracting the internal offsets is as shown in the table below with the worst case internal offsets being ±17 mV.(See Table 5) Offset Range (Reg 06h bit 0) 1 0 Max Offset Blk Acc=511 ~30 mV ~11 mV Min Offset Blk Acc=0 ~-72 mV ~-40 mV Accumulator LSB Size ~0.2 mV ~0.1 mV Table 5. Black Level Control - General The black loop is a feedback system that causes the ADC output to settle to 64 during the register defined black pixels. This has the purpose of removing any CCD and system offsets and defining 64 as the known black level. The loop has an exponential settling response and the time constant of this loop is effected by the black loop gain and the number of black pixels to accumulate before updating the black accumulator. See Figure 10 for a block diagram of the black level loop. DS322PP1 21 CS7622 Black Level Control - Loop Gain, Clamp Length Default = 2Ah; Read/Write (address 0Dh) Bit Number Bit Name Default Bit 7:6 5:0 7 6 5 4 3 2 1 0 blk_gain1 blk_gain0 blk_clp_15 blk_clp_14 blk_clp_13 blk_clp_12 blk_clp_11 blk_clp_10 0 0 1 0 1 0 1 0 Mnemonic Function blk_gain1-0 Black Loop Gain Factor: can be set to 1x,2x,4x,or 8x and is simply a multiplying constant to effect the weight of each black pixel before it is accumulated. 00 - defines a gain of 1x 01 - defines a gain of 2x 10 - defines a gain of 4x 11 - defines a gain of 8x blk_clp_15-10 Black Loop Clamp Length: The black clamp length effects the loop time constant and also acts to average out noise in the black level. The larger this value the more pixels that are summed before the loop is updated which causes greater averaging and a smaller settling time constant. The table below shows the black loop time constant for various settings of Offset Range (register 04h, bit 0) and Fixed Gain Settings (register 14h, bits 5-3). (See Table 5) Fixed Gain (Register 16h) not fixed x1 x2 x4 x8 Offset Range = 1 Offset Range = 0 -1/(ln(1-nK))(1/fu) -1/(ln(1-nK/8))(1/fu) -1/(ln(1-nK/4))(1/fu) -1/(ln(1-nK/2))(1/fu) -1/(ln(1-nK))(1/fu) -1/(ln(1-nK/2))(1/fu) -1/(ln(1-nK/16))(1/fu) -1/(ln(1-nK/8))(1/fu) -1/(ln(1-nK/4))(1/fu) -1/(ln(1-nK/2))(1/fu) Table 6. Where: K = 1/256*blk_gain n = Black loop clamp length = blk_clp_l[5:0] fu = update rate 22 DS322PP1 CS7622 Gain Calibration Offset 1 Default = 00h; Read only (address 0Eh) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset 17 16 15 14 13 12 11 10 0 0 0 0 0 0 0 0 Bit Mnemonic Function 7:0 gain_offset17-10 offset added to 4x gain segment, values are in 2’s complement. See details in register 10h. Gain Calibration Offset 2 Default = 00h; Read only (address 0Fh) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset 27 26 25 24 23 22 21 20 0 0 0 0 0 0 0 0 Bit Mnemonic Function 7:0 gain_offset27-20 offset added to 2x gain segment, values are in 2’s complement. See details in register 10h. Gain Calibration Offset 3 Default = 00h; Read only (address 10h) Bit Number Bit Name Default Bit 7:0 DS322PP1 7 6 5 4 3 2 1 0 gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset gain_offset 37 36 35 34 33 32 31 30 0 0 0 0 0 0 0 0 Mnemonic Function gain_offset37-30 Offset added to 1x gain segment. Values are in 2’s complement. These registers are used to report some of the calibration settings. After calibration is performed the gain offset registers are automatically updated with values needed for the DRX circuitry to operate correctly. These registers should not be written to since this will remove the proper settings found during calibration. The gain offset values are used to add an offset to the output of the ADC when using different analog gain settings (See equations below). The purpose of this is to produce a continuous transition between the different gain settings so that the final 13 bit output is monotonic and has no undesired artifacts. (See Figure 17) {ADC_out if in the 8x gain segment} dout[12:0] = {ADC_out*2+Offset1 if in the 4x gain segment} {ADC_out*4+Offset2*2 if in the 2x gain segment} {ADC_out*8+Offset3*4 if in the 1x gain segment} 23 CS7622 ADC OUT 1024 512 8X 4X 2X 1X 64 4096 USE OFFSET3 USE OFFSET1 8192 USE OFFSET2 INPUT 2048 1024 64 1.0 INPUT Figure 17. Transfer Function of Analog Input to Digital Output (assuming full scale level of 1.0 V) 24 DS322PP1 CS7622 Fixed Gain Default = 00h; Read/Write (address 14h) Bit Number Bit Name Default 7 RESERVED 0 Bit Mnemonic 7:6, 2:0 - 5:3 DS322PP1 6 fixed_gain2-0 5 4 3 2 fixed_gain2 fixed_gain1 fixed_gain0 0 0 0 0 1 0 RESERVED 0 0 0 Function Reserved Fixed Gain: This is used to turn off the DRX functionality and apply a fixed gain to the input before reaching the ADC. A setting of 000 is used for normal operation this will yield the largest dynamic range by switching the front end gain relative to the amplitude of the input signal. The settings 001, 010, 011, and 100 are for fixed gains of 1x, 2x, 4x, and 8x respectively. Figure 18 shows the transfer function of the output of the ADC for a given input with the various fixed gain settings. 25 CS7622 FIXED GAIN = 000 ADC OUTPUT 1024 8X 4X 2X 1X INPUT 0.125 0.25 ADC OUTPUT 0.5 1.0 FIXED GAIN = 001 1024 1X 1.0 ADC OUTPUT INPUT (V) FIXED GAIN = 010 1024 2X INPUT (V) 0.5 FIXED GAIN = 011 ADC OUTPUT 1.0 1024 4X INPUT (V) ADC OUTPUT 0.25 0.5 FIXED GAIN = 100 1.0 0.25 1.0 1024 8X INPUT (V) 0.125 0.5 Figure 18. Transfer Function of ADC with Fixed Gain Settings (assuming full scale level of 1.0 V) 26 DS322PP1 CS7622 Compander - Black slope, Slopes (MSBs) Default = 10h; Read/Write (address 15h) Bit Number Bit Name Default 7 6 5 RESERVED 0 Bit Mnemonic 7:5 - 0 0 4 3 2 1 0 comp_linear slope18 slope28 slope38 slope48 1 0 0 0 0 Function Reserved Compander Black Level Slope: 0 - The values of “0” to “64” in a 13 bit representation are set to “offset1” in a 10 bit representation. Offset1 can be set in register 33h. 1 - In this case the black level is mapped linearly from 13 bit values to 10 bit values. “64” is mapped into “8”. All the other values between “0” and “64” are divided by 8 in order to get the 10 bit representation. (See Figure 13) 4 comp_linear 3 slope18 Compander Slope 1: MSB of slope of first segment of companding curve. (See Figure 13) 2 slope28 Compander Slope 2: MSB of slope of second segment of companding curve. (See Figure 13) 1 slope38 Compander Slope 3: MSB of slope of third segment of companding curve. (See Figure 13) 0 slope48 Compander Slope 4: MSB of slope of fourth segment of companding curve. (See Figure 13) Compander Slope 1 (LSBs) Default = B2h; Read/Write (address 16h) Bit Number Bit Name Default 7 6 3 2 1 0 slope16 5 slope15 4 slope17 slope14 slope13 slope12 slope11 slope10 1 0 1 1 0 0 1 0 Bit Mnemonic 7:0 slope17-10 Function Compander - Slope1: Slope of first segment (slope1[8:0]) of companding curve. Max value is 1.996. The LSB step size is 0.0039. (See Figure 13) Compander Slope 2 (LSBs) Default = 60h; Read/Write (address 17h) Bit Number Bit Name Default 7 6 3 2 1 0 slope26 5 slope25 4 slope27 slope24 slope23 slope22 slope21 slope20 0 1 1 0 0 0 0 0 Bit Mnemonic Function 7:0 slope27-20 Compander - Slope2: Slope of second segment (slope2[8:0]) of companding curve. Max value is 1.996. The LSB step size is 0.0039. (See Figure 13) DS322PP1 27 CS7622 Compander Slope 3 (LSBs) Default = 20h; Read/Write (address 18h) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 slope37 slope36 slope35 slope34 slope33 slope32 slope31 slope30 0 0 2 0 0 0 0 0 Bit Mnemonic 7:0 slope37-30 Function Compander - Slope3: Slope of third segment (slope3[8:0]) of companding curve. Max value is 1.996. The LSB step size is 0.0039. (See Figure 13) Compander Slope 4 (LSBs) Default = 07h; Read/Write (address 19h) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 slope47 slope46 slope45 slope44 slope43 slope42 slope41 slope40 0 0 0 0 0 1 1 1 Bit Mnemonic 7:0 slope47-40 Function Compander - Slope4: Slope of fourth segment (slope4[8:0]) of companding curve. Max value is 1.996. The LSB step size is 0.0039. (See Figure 13) Compander Offset 1 Default = 08h; Read/Write (address 1Ah) Bit Number Bit Name Default 28 7 6 3 2 1 0 offset16 5 offset15 4 offset17 offset14 offset13 offset12 offset11 offset10 0 0 0 0 1 0 0 0 Bit Mnemonic 7:0 offset17-10 Function Compander - Offset1: Black level value of companding curve if not in linear mapping mode (comp_linear = 0). (See Figure 13) DS322PP1 CS7622 Compander Offset 2 (MSBs) Default = 0Bh; Read/Write (address 1Bh) Bit Number Bit Name Default 7 6 RESERVED 0 5 4 3 2 1 0 offset29 offset28 offset39 offset38 offset49 offset48 0 0 1 0 1 1 0 Bit Mnemonic Function 7:6 - 5:4 offset29-28 MSBs of offset of second segment of companding curve. (See Figure 13) 3:2 offset39-38 MSBs of offset of third segment of companding curve. (See Figure 13) 1:0 offset49-48 MSBs of offset of fourth segment of companding curve. (See Figure 13) Reserved Compander Offset 2 (LSBs) Default = BFh; Read/Write (address 1Ch) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 offset27 offset26 offset25 offset24 offset23 offset22 offset21 offset20 1 0 1 1 1 1 1 1 Bit Mnemonic Function 7:0 offset27-20 Offset of second segment (offset2[9:0]) of companding curve. (See Figure 13) Compander Offset 3 (LSBs) Default = 05h; Read/Write (address 1Dh) Bit Number Bit Name Default 7 6 3 2 1 0 offset36 5 offset35 4 offset37 offset34 offset33 offset32 offset31 offset30 0 0 0 0 0 1 0 1 Bit Mnemonic 7:0 offset37-30 Function Offset of third segment (offset3[9:0]) of companding curve. (See Figure 13) Compander Offset 4 (LSBs) Default = 20h; Read/Write (address 1Eh) Bit Number Bit Name Default 7 6 offset47 0 Bit Mnemonic 7:0 offset47-40 DS322PP1 4 3 2 1 0 offset46 5 offset45 offset44 offset43 offset42 offset41 offset40 0 1 0 0 0 0 0 Function Offset of fourth segment (offset4[9:0]) of companding curve. (See Figure 13) 29 CS7622 Compander X1 (MSBs) Default = 03h; Read/Write (address 1Fh) Bit Number Bit Name Default 7 6 5 RESERVED 0 Bit Mnemonic 7:5 - 4:0 x112-x18 0 0 4 3 2 1 0 x112 x111 x110 x19 x18 0 0 0 1 1 Function Reserved End value of first segment of companding curve (MSBs). (See Figure 13) Compander X1 (LSBs) Default = 20h; Read/Write (address 20h) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 x17 x16 x15 x14 x13 x12 x11 x10 0 0 1 0 0 0 0 0 Bit Mnemonic 7:0 x17x10 Function End value of first segment (x1[12:0]) of companding curve (LSBs). (See Figure 13) Compander X2 (MSBs) Default = 05h; Read/Write (address 21h) Bit Number Bit Name Default 7 6 5 RESERVED 0 Bit Mnemonic 7:5 - 4:0 x212-x28 0 0 4 3 2 1 0 x212 x211 x210 x29 x28 0 0 1 0 1 Function Reserved End value of second segment of companding curve (MSBs). (See Figure 13) Compander X2 (LSBs) Default = 18h; Read/Write (address 22h) Bit Number Bit Name Default 30 7 6 5 4 3 2 1 0 x27 x26 x25 x24 x23 x22 x21 x20 0 0 0 1 1 0 0 0 Bit Mnemonic 7:0 x27-x20 Function End value of second segment (x2[12:0]) of companding curve (LSBs). (See Figure 13) DS322PP1 CS7622 Compander X3 (MSBs) Default = 0Bh; Read/Write (address 23h) Bit Number Bit Name Default 7 6 5 RESERVED 0 Bit Mnemonic 7:5 - 4:0 x312-x38 0 0 4 3 2 1 0 x312 x311 x310 x39 x38 0 1 0 1 1 Function Reserved End value of third segment of companding curve (MSBs). (See Figure 13) Compander X3 (LSBs) Default = 58h; Read/Write (address 24h) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 x37 x36 x35 x34 x33 x32 x31 x30 0 1 0 1 1 0 0 0 Bit Mnemonic 7:0 x37-x30 Function End value of third segment (x3[12:0]) of companding curve (LSBs). (See Figure 13) Device ID Default = CCh; Read only (address 25h) Bit Number Bit Name Default 7 6 5 4 3 2 1 0 device_ID7 device_ID6 device_ID5 device_ID4 device_ID3 device_ID2 device_ID1 device_ID0 1 Bit Mnemonic 7:0 device_ID7-0 0 0 0 1 1 0 0 Function This read-only register is the unique ID for the CS7622. Revision Code Default = 00h; Read only (address 26h) Bit Number Bit Name Default 7 rev_code7 6 5 4 3 2 1 0 rev_code6 rev_code5 rev_code4 rev_code3 rev_code2 rev_code1 rev_code0 0 0 0 0 0 0 0 0 Bit Mnemonic 7:0 rev_code7-0 DS322PP1 Function This read-only register is the revision code for the CS7622. 31 CS7622 5.0 PIN DESCRIPTIONS GNDD DOUT6 VDDD DOUT5 DOUT7 DOUT4 DOUT8 DOUT3 DOUT9 DOUT10 DOUT11 24 2 23 3 DOUT12 4 SCLK 5 SDATO 6 SDATI SEN TEST DOUT2 32 31 30 29 28 27 26 25 1 CS7622 32-pin TQFP Top View 7 8 22 21 CLKO 20 CK_DATA 19 CK_FT 18 9 10 11 12 13 14 15 16 DOUT1 DOUT0 17 CLAMP RST DIAG BG_RES REF_CAPP AIN GNDA REF_CAPN VDDA Supply VDDA - Supply for analog Pin 13 3.3 V analog supply. VDDD - Supply for digital Pin 28 3.3 V or 2.7 V digital supply. Ground GNDA - Ground for analog Pin 12 GNDA is supplied by VDDA. GNDD - Ground for digital Pin 27 Supplied by VDDD. CMOS Input CLAMP - Black level clamp signal Pin 18 Provided by the external timing generator. Supplied by VDDD. CK_FT - Clock in feed-through Pin 19 Sampling clock for feed-through level. CK_DATA - Clock in data Pin 20 Sampling clock for data level. Supplied by VDDD REF_CAPN - Reference capacitor- negative terminal Pin 14 32 Supplied by VDDA. A 1 µF ceramic capacitor should be connected between REF_CAPN and REF_CAPP. DS322PP1 CS7622 REF_CAPP - Reference capacitor- positive terminal Pin 15 Supplied by VDDA. A 1 µF ceramic capacitor should be connected between REF_CAPN and REF_CAPP. RST - Reset pin, negative true Pin 17 May be connected to external power-on-reset-circuit. Supplied by VDDD. SCLK - Serial bus clock signal Pin 5 Supplied by VDDD. SDATI - Serial bus data input signal Pin 7 Supplied by VDDD. SEN - Serial bus enable signal-chip select (active low) Pin 8 Supplied by VDDD. TEST - Test enable pin Pin 9 Supplied by VDDD. CMOS Analog Input AIN - Video data input from CCD Pin 11 Supplied by VDDA. BG_RES - Band-gap resistor Pin 10 Supplied by VDDA. A 10 kΩ resistor should be connected between BG_RES and GNDA. CMOS 4 mA Output CLKO - Clock = output Pin 21 Signal on this pin can either be the pixel clock output or data_valid signal output. Supplied by VDDD. DOUT[0:12] - Digitized CCD data output Pins 22-32, and 1-4 DOUT0 is LSB. Supplied by VDDD. SDATO - Serial bus data output signal Pin 6 DS322PP1 Supplied by VDDD. 33 CS7622 6.0 PACKAGE DIMENSIONS 32L TQFP PACKAGE DRAWING E E1 D D1 1 e B ∝ A A1 L INCHES MIN MAX --0.063 0.002 0.006 0.012 0.018 0.343 0.366 0.272 0.280 0.343 0.366 0.272 0.280 0.028 0.035 0.018 0.030 ∝ 0.000° 7.000° * Nominal pin pitch is 0.50 mm DIM A A1 B D D1 E E1 e* L MILLIMETERS MIN MAX --1.60 0.05 0.15 0.30 0.45 8.70 9.30 6.90 7.10 8.70 9.30 6.90 7.10 0.70 0.90 0.45 0.75 0.00° 7.00° Controlling dimension is mm. JEDEC Designation: MS026 34 DS322PP1 • Notes •