LM9627 Color CMOS Image Sensor VGA 30 FPS General Description Applications The LM9627 is a high performance, low power, third inch VGA CMOS Active Pixel Sensor capable of capturing color digital still or motion images and converting them to a digital data stream. * * * * * * In addition to the active pixel array, an on-chip 12 bit A/D convertor, fixed pattern noise elimination circuits and a video gain amplifier is provided. Furthermore, an integrated programmable smart timing and control circuit allows the user maximum flexibility in adjusting integration time, active window size, gain and frame rate. Various control, timing and power modes are also provided. PC Camera Digital Still Camera Video Conferencing Security Cameras Toys Machine Vision Key Specifications Total: 664H x 504V Active: 648H x 488V * Array Format * Effective Image Area Features * Optical Format * * * * * * * * * * * Pixel Size Supplied with micro lenses Video or snapshot operations Programmable pixel clock, inter-frame and inter-line delays. Programmable partial or full frame integration Programmable gain adjustment Horizontal & vertical sub-sampling (2:1 & 4:2) Windowing External snapshot trigger & event synchronisation signals Auto black level compensation Flexible digital video read-out supporting programmable: - polarity for synchronisation and pixel clock signals - leading edge adjustment for horizontal synchronization * Programmable via 2 wire I2C compatible serial interface * Power on reset & power down mode Total: 4.98mm x 3.78 mm Active: 4.86 mm x 3.66 mm 1/3" 7.5m x 7.5m * Video Outputs 8,10 & 12 Bit Digital * Dynamic Range 57dB * FPN 0.35% * Sensitivity red green blue 14.5 kLSBs/lux.s 7.5 kLSBs/lux.s 5.1 kLSBs/lux.s * Quantum Efficiency 27% * Fill Factor 47% (no micro lens) * Color Mosaic Bayer pattern * Package 48 LCC * Single Supply 3.3 V * Power Consumption 90 mW 0 to 50o C * Operating Temp System Block Diagram Storage lens LM9627 12bit digital image Digital Image Processor I2 C compatible event trigger snapshot 2000 National Semiconductor Corporation Confidential www.national.com LM9627 Color CMOS Image Sensor VGA 30 FPS March 2001 Digital Video Framer 12 Bit A/D AMP Black Level Compensation APS Array Bad Pixel Detect & Correct Horizontal Shift Register Column CDS Row Address Decoder d[11:0] pclk hsync vsync POR Reset Gen Vertical Timing Row Address Gen Horizontal Timing Gain Control Register Bank Clock Gen Controller (sequencer) Power Control Master Timer mclk I2C Compatible Serial I/F sda sclk sadr snapshot irq pdwn Figure 1. Chip Block Diagram extsync vrl vsrvdd extsync 6 5 4 3 2 1 48 47 46 45 44 43 NC vdd_pix vdd_od3 irq vss_od3 sadr vss_od1 sda vdd_od1 Connection Diagram sclk 7 42 NC snapshot 8 41 fine_i resetb 9 40 gnd pdwn 10 39 fine_ctrl 38 offset 37 vdd_ana1 vss_ana1 vss_dig 11 vdd_dig 12 hsync 13 36 vsync 14 35 vref_adc LM9627 48 PIN LCC 31 19 20 21 22 23 24 25 26 27 28 29 30 NC 18 d11 vss_od2 vdd_od2 d10 32 d9 17 d8 d0 d7 vdd_ana2 d6 33 d5 16 d4 mclk d3 34 d2 15 d1 pclk vss_ana2 NC LM9627 Overall Chip Block Diagram Ordering Information Temperature (0C TA +50C) LM9627 CCEA Confidential NS Package LCC 2 www.national.com LM9627 Typical Application Circuit 37 vdd_ana1 0.1F 7 5 6 sadr 48 sda irq 8 Serial Control Bus sclk 4 snapshot 3.3V analog 10 resetb mclk 9 pdwn 16 Camera Control extsync System Control 3.3V analog vdd_ana2 33 36 vss_ana1 vss_ana2 34 47 vdd_od1 vdd_od2 31 46 vss_od1 vss_od2 32 44 vdd_od3 vdd_dig 45 vss_od3 vss_dig 11 3.3V digital 0.1F 3.3V digital 0.1F 3.3V analog 0.1F 0.1F 3.3V digital 0.1F 3.3Vdigital 3 vdd_pix LM9627 12 0.1F vsrvdd 1 1.0F 2 vrl 3.3V analog vdd_ana 1.5k 35 820 vref_adc 2 2 k 1% fine_i 41 fine_ctrl 39 0.1F 1N4148 2N3904 1 0 k 1% NC NC NC NC 1 . 2 k 1% offset 38 470 1% 13 14 15 d0 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10 d11 gnd 40 pclk vsync 4.7F hsync 18 19 42 43 vdd_ana 30 29 28 27 26 25 24 23 22 21 20 17 Digital Video Bus Figure 2. Typical Application Diagram Scan Read Out Direction pin 1 vertical scan (0,0) (0,0) digital out (0,0) horizontal scan lens CMOS Image Sensor Figure 3. Scan directions and position of origin in imaging system Confidential 3 www.national.com LM9627 Pin Descriptions Pin Name I/O Typ Description 1 vsrvdd I0 P Analog bidirectional, it should be connect to ground via a 1.0f capacitor. This pin is the internal charge pump voltage source. 2 vrl I A Anti blooming pin. This pin is normally tied to ground. 3 vdd_pix I P 3.3 volt supply for the pixel array. 4 irq O D Digital output, the interrupt request pin. This pin generates interrupts during snapshot mode. 5 sadr I D Digital input with pull down resistor. This pin is used to program different slave addresses for the sensor in an I2 C compatible system. 6 sda IO D I2 C compatible serial interface data bus. The output stage of this pin has an open drain driver. 7 sclk I D I2 C compatible serial interface clock. 8 snapshot I D Digital input with pull down resistor used to activate (trigger) a snapshot sequence. 9 resetb I D Digital input with pull up resistor. When forced to a logic 0 the sensor is reset to its default power up state. The resetb signal is internally synchronized to mclk which must be running for a reset to occur. 10 pdwn I D Digital input with pull down resistor. When forced to a logic 1 the sensor is put into power down mode. 11 vss_dig I P 0 volt power supply for the digital circuits. 12 vdd_dig I P 3.3 volt power supply for the digital circuits. D Digital Bidirectional. This is a dual mode pin. When the sensor's digital video port is configured to be a master, (the default), this pin is an output and is the horizontal synchronization pulse. When the sensor's digital video port is configured to be a slave, this pin is an input and is the row trigger. 13 hsync IO 14 vsync IO D Digital Bidirectional. This is a dual mode pin. When the sensor's digital video port is configured to be a master, (the default), this pin is an output and is the vertical synchronization pulse. When the sensor's digital video port is configured to be a slave, this pin is an input and is the frame trigger. 15 pclk O D Digital output. The pixel clock. 16 mclk I D Digital input. The sensor's master clock input. 17 d0 O D Digital output. Bit 0 of the digital video output bus. This output can be put into tri-state mode. 18 NC Pin not used, do not connect. 19 NC Pin not used, do not connect. 20 d1 O D Digital output. Bit 1 of the digital video output bus. This output can be put into tri-state mode. 21 d2 O D Digital output. Bit 2 of the digital video output bus. This output can be put into tri-state mode. 22 d3 O D Digital output. Bit 3 of the digital video output bus. This output can be put into tri-state mode. 23 d4 O D Digital output. Bit 4 of the digital video output bus. This output can be put into tri-state mode. 24 d5 O D Digital output. Bit 5 of the digital video output bus. This output can be put into tri-state mode. 25 d6 O D Digital output. Bit 6 of the digital video output bus. This output can be put into tri-state mode. Confidential 4 www.national.com LM9627 Pin Descriptions (Continued) Pin Name I/O Typ Description 26 d7 O D Digital output. Bit 7 of the digital video output bus. This output can be put into tri-state mode. 27 d8 O D Digital output. Bit 8 of the digital video output bus. This output can be put into tri-state mode. 28 d9 O D Digital output. Bit 9 of the digital video output bus. This output can be put into tri-state mode. 29 d10 O D Digital output. Bit 10 of the digital video output bus. This output can be put into tri-state mode. 30 d11 O D Digital output. Bit 11 of the digital video output bus. This output can be put into tri-state mode. 31 vdd_od2 I P 3.3 volt supply for the digital IO buffers. 32 vss_od2 I P 0 volt supply for the digital IO buffers 33 vdd_ana2 I P 3.3 volt supply for analog circuits. 34 vss_ana2 I P 0 volt supply for analog circuits. 35 vref_adc I A A/D reference resistor ladder voltage. See figure 4 for equivalent circuit. 36 vss_ana1 I P 0 volt supply for analog circuits. 37 vdd_ana1 I P 3.3 volt supply for analog circuits. 38 offset I A Analog input used to adjust the offset of the sensor. See figure 4 for equivalent circuit. 39 fine_ctrl O A Analog output used to drive the offset pin. 40 gnd 41 fine_i 42 NC Pin not used, do not connect. 43 NC Pin not used, do not connect. 44 vdd_od3 I P 3.3 volt supply for the sensor. 45 vss_od3 I P 0 volt supply for the sensor. 46 vss_od1 I P 0 volt supply for the digital IO buffers 47 vdd_od1 I P 3.3 volt supply for the digital IO buffers. 48 extsync O D Digital output. The external event synchronization signal is used to synchronize external events in snapshot mode. This pin must be tied to ground. I A Bias current for the fine offset adjust. Legend: (I=Input), (O=Output), (IO=Bi-directional), (P=Power), (D=Digital), (A=Analog). adc_vref offset 800 1K 200 Figure 4. Equivalent Circuits For adc_ref and offset pins Confidential 5 www.national.com LM9627 Absolute Maximum Ratings (Notes 1 & 2) Operating Ratings Any Positive Supply Voltage 6.5V Voltage On Any Input or Output Pin -0.5V to 6.5V Input Current at any pin (Note 3) 25mA ESD Susceptibility (Note 5) Human Body Model 2000V Machine Model 200V Package Input Current (Note 3) 50mA Package Power Dissipation @ TA (Note 4) 2.5W Soldering Temperature Infrared, 10 seconds (Note 6) 220C Storage Temperature -40C to 125C Operating Temperature Range All VDD Supply Voltages Voltage Range on vref_adc pin Voltage Range on offset pin (Notes 1 & 2) 0CT+50C +3.15V to +3.6V +0.6V to +1.0V +0.04V to +0.4V DC and logic level specifications The following specifications apply for all VDD pins= +3.3V. Boldface limits apply for TA = TMIN to TMAX : all other limits TA = 25o C (Note 7) Symbol Parameter Conditions Min note 9 Typical note 8 Max note 9 Units sclk, sda, sadr, Digital Input/Output Characteristics VIH Logical "1" Input Voltage 0.7* vdd_od vdd_od+0.5 V VIL Logical "0" Input Voltage -0.5 0.3* vdd_od V VOL Logical "0" Output Voltage vdd_od = +3.15V, Iout=3.0mA 0.5 V Vhys Hysteresis (SCLK pin only) vdd_od = +3.15V Ileak Input Leakage Current Vin=vss_od 0.05*vdd_o d V -1 mA mclk, snapshot, pdwn, resetb, hsync, vsync Digital Input Characteristics VIH Logical "1" Input Voltage vdd_dig = +3.6V 2.0 V VIL Logical "0" Input Voltage vdd_dig = +3.15V IIH Logical "1" Input Current VIH = vdd_dig 0.1 mA IIL Logical "0" Input Current VIL = vss_dig -1 mA 0.8 V d0 - d11, pclk, hsync, vsync, extsync, irq, Digital Output Characteristics VOH Logical "1" Output Voltage vdd_od=3.15V, Iout=-1.6mA VOL Logical "0" Output Voltage vdd_od=3.15V, Iout =-1.6mA IOZ TRI-STATE Output Current VOUT = vss_od VOUT = vdd_od IOS Output Short Circuit Current 2.2 V 0.5 V -0.1 0.1 mA mA +/-17 mA Power Supply Characteristics IA Analog Supply Current Power down mode, no clock. Operational mode in dark 700 19 mA mA ID Digital Supply Current Power down mode, no clock. Operational mode in dark 300 7 mA mA Power Dissipation Specifications The following specifications apply for All VDD pins = +3.3V. Boldface limits apply for TA = T MIN to TMAX : all other limits TA = 25 o C. Symbol Parameter Conditions Min note 9 Typical note 8 Max note 9 Units Pdwn Power Down no clock running 5 mW PWR Average Power Dissipation mclk = 48Mhz & sensors default settings in dark. 90 mW Confidential 6 www.national.com The following specifications apply for all VDD pins= +3.3V. Boldface limits apply for TA = TMIN to TMAX : all other limits TA = 25oC. Symbol Parameter Min note 9 Conditions Video Amplifier Nominal Gain 64 linear steps Typical note 8 Max note 9 Units 0-15 dB AC Electrical Characteristics The following specifications apply for All VDD pins = +3.3V. Boldface limits apply for T A = TMIN to T MAX: all other limits TA = 25 o C. Symbol Fmclk Parameter Conditions Input Clock Frequency Min note 9 Max note 9 Units 12 48 MHz Tch Clock High Time @ CLKmax 10 45 ns Tcl Clock Low Time @ CLKmax 10 45 ns Clock Duty Cycle @ CLKmax 45/55 55/45 min/max Trc , Tfc Clock Input Rise and Fall Time 3 Fhclk Internal System Clock Frequency 1.0 Treset Reset pulse width 1.0 FRM rate Note 1: Note 2: Note 3: Note 4: Typical note 8 Frame Rate ns 14.0 MHz s 1 30 fps Absolute Maximum Ratings indicate limits beyond which damage to the device may occur. Operating Ratings indicate conditions for which the device is functional, but do not guarantee specific performance limits. For guaranteed specifications and test conditions, see the Electrical Characteristics. The guaranteed specifications apply only for the test conditions listed. Some performance characteristics may degrade when the device is not operated under the listed test conditions. All voltages are measured with respect to VSS = vss_ana = vss_od = vss_dig = 0V, unless otherwise specified. When the voltage at any pin exceeds the power supplies (VIN < VSS or VIN > VDD), the current at that pin should be limited to 25mA. The 50mA maximum package input current rating limits the number of pins that can safely exceed the power supplies with an input current of 25mA. The absolute maximum junction temperature (TJmax) for this device is 125 o C. The maximum allowable power dissipation is dictated by TJmax, the junction-to-ambient thermal resistance (JA), and the ambient temperature (TA), and can be calculated using the formula PDMAX = (TJmax - TA)/JA. In the 48-pin LCC, JA is 38.5 o C/W, so PDMAX = 2.5W at 25o C Note 5: Note 6: Note 7: and 1.94W at the maximum operating ambient temperature of 50oC. Note that the power dissipation of this device under normal operation will be well under the PDMAX of the package. Human body model is 100pF capacitor discharged through a 1.5k resistor. Machine model is 220pF discharged through ZERO Ohms. See AN450, "Surface Mounting Methods and Their Effect on Product Reliability", or the section entitled "Surface Mount" found in any post 1986 National Semiconductor Linear Data Book, for other methods of soldering surface mount devices. The analog inputs are protected as shown below. Input voltage magnitude up to 500mV beyond the supply rails will not damage this device. However, input errors will be generated If the input goes above AV+ and below AGND. VDD Pad Internal Circuits IOP VSS Note 8: Note 9: Typical figures are at TJ = 25o C, and represent most likely parametric norms. Test limits are guaranteed to National's AOQL (Average Outgoing Quality Level). Confidential 7 www.national.com LM9627 Video Amplifier Specifications LM9627 CMOS Active Pixel Array Specifications Parameter Value Number of pixels (column, row) Total Active 664 x 504 648 x 488 Array size (x,y Dimensions) Total Active Units pixels pixels 4.98 x 3.78 4.86 x 3.66 mm mm Pixel Pitch 7.5 Fill Factor (without micro-lens) 47 % Image Sensor Specifications The following specifications apply for All VDD pins = +3.3V, TA = 25o C, Illumination Color Temperature = 2850 o K, IR cutoff filter at 700nm, mclk = 48MHz, frame rate = 30Hz, vref_adc = 0.6 volt, video gain 0dB. Parameter Min Conditions Optical Sensitivity @ A/D output red green blue Typical note 1 14.5 7.5 5.1 Optical Sensitivity @ A/D input red green blue Max Units kLSBs/(lux.s) 2.12 1.1 0.75 volt/(lux.s) Dynamic Range 57 dB Read Noise 5.3 LSBs Offset Fixed Pattern Noise RMS value of pixel FPN in dark as a percentage of full scale. 0.35 % Sensitivity Fixed Pattern Noise RMS variation of pixel sensitivities as a percentage of the average sensitivity. 1 % Note 1: Typical figures are at TJ = 25o C, and represent most likely parametric norms. Confidential 8 www.national.com LM9627 Sensor Response Curves 8.00E+02 red Spectral sensitivity [V/((W/m^2)*s)] 7.00E+02 6.00E+02 green 5.00E+02 blue 4.00E+02 3.00E+02 2.00E+02 1.00E+02 0.00E+00 370 420 470 520 570 620 670 720 770 820 wavelength [nm] 5000 4000 ADC output code [LSBs] A/D output code Figure 5. Spectral Response Curve 3000 green blue 2000 red 1000 0 0 0.1 0.2 0.3 0.4 0.5 0.6 Exposure [lux.s] Figure 6. Linearity Response Curve Confidential 9 www.national.com 1.0 OVERVIEW 0-15dB 1.1 Light Capture and Conversion The LM9627 contains a CMOS active pixel array consisting of 648 rows by 488 columns. This active region is surrounded by 8 columns and 8 rows of optically shielded (black) pixels as shown in Figure7. 648 columns, 488 rows color (Bayer pattern) active pixels Video AMP 12 Bit A/D Analog pixel values 8 columns, 8 rows black pixels Digital pixel data Figure 9: Analog Signals Conditioning & Conversion to Digital do[11:0] Digital Video Framer 8 columns, 8 rows black pixels Black Level Compensation Bad Pixel Correction The digital pixel data is further processed to: * remove defects due to bad pixels, * compensate black level, before being framed and presented on the digital output port. (see Figure 10). pclk hsync vsync Figure 10. Digital Pixel Processing. Figure 7: CMOS APS region of the LM9627 The color filters are Bayer pattern coded starting at row 8 and column 8. (rows 0 to 7 & columns 0 to 7 are black). The color coding is green, red, green, red until the end of row 8, then blue, green, blue, green until the end or row 9 and so on (see Figure 7). 1.2 Program and Control Interfaces The programming, control and status monitoring of the LM9627 is achieved through a two wire I2 C compatible serial bus. In addition, a slave address pin is provided (see Figure 11). At the beginning of a given integration time the on-board timing and control circuit will reset every pixel in the array one row at a time as shown in Figure 8. Note that all pixels in the same row are simultaneously reset, but not all pixels in the array. a b c Line Address LM9627 Functional Description d e f g h i j k sda Register Bank I2C Compatible Serial I/F sclk l m n o p q r sadr 0 1 2 3 4 5 6 7 8 9 10 11 Figure 11. Control Interface to the LM9627. Additional control and status pins: snapshot and external event synchronization are provided allowing the latency of the serial control port to be bypassed during single frame capture. An interrupt request pin is also available allowing complex snapshot operations to be controlled via an external micro-processor (see Figure 12). 12 13 14 15 Analog Data Out CDS/Shift Register irq Timing Generator Figure 8: CMOS APS Row and Column addressing scheme At the end of the integration time, the timing and control circuit will address each row and simultaneously transfer the integrated value of the pixel to a correlated double sampling circuit and then to a shift register as shown in Figure 8. extsyn snapshot Figure 12. Snapshot & External Event Trigger Signals Once the correlated double sampled data has been loaded into the shift register, the timing and control circuit will shift them out one pixel at a time starting with column "a". The pixel data is then fed into an analog video amplifier, where a user programmed gain is applied (see Figure 9). After gain adjustment the analog value of each pixel is converted to a 12 bit digital data as shown in Figure 9. Confidential 10 www.national.com 2.0 Column/Horizontal a b c d e f g h i j k l mn o p q r WINDOWING The integrated timing and control circuit allows any size window in any position within the active region of the array to be read out with a 1x1 pixel resolution. The window read out is called the "Display Window". Row/Vertical A "Scan Window" must be defined first, by programing the start and end row addresses as shown in Figure 13. Four coordinates (start row address, start column address, end row address & end column address) are programmed to define the size and location of the "Display Window" to be read out (see Figure 13). display col display col scan row end address start address start address display row start address display row end address Display Window scan row end address Figure 14: Progressive Scan Read Out Mode Scan Window Active Pixel Array Figure 13. Windowing Notes: * The "Display Window" must always be defined within the "Scan Window". * A "Display Window" can only be read out in the progressive scan mode. * By default the "Display Window" is the complete array. 3.2 Interlaced Readout Mode In interlaced readout mode, pixels are read out in two fields, an Odd Field followed by an Even Field. The Odd Field, consisting of all even row pairs contained within the display window, is read out first. Each pixel in the "Odd Field" is consecutively read out, one pixel at a time, starting with the left most pixel in the top most row pair. The Even Field, consisting of all odd row pairs contained within the display window, is then read out. Each pixel in the "Even Field" is consecutively read out, one pixel at a time, starting with the left most pixel in the top most row pair. 2.1 Programming the scan window Two registers (SROWS & SROWE) are provided to program the size of the scan window. The start and end row address of the scan window is given by: Row/Vertical Column/Horizontal a b c d e f g h i j k l mn o p q r scan row start address = (2* SwStartRow) + SwLsb scan row end address = (2* SwEndRow) + 1 + SwLsb Where: SwStartRow is the contents of the Scan Window start row register (SROWS) SwEndROW is the contents of the Scan Window end row register (SROWE) SwLsb is bit 6 of the Display Window LSB register (DWLSB) READ OUT MODES 3.1 Progressive Scan Readout Mode In progressive scan readout mode, every pixel in every row in the display window is consecutively read out, one pixel at a time, starting with the left most pixel in the top most row. Hence, for the example shown in Figure 14, the read out order will be a0,b0,...,r0 then a1,b1,...,r1 and so on until pixel r20 is read out. Confidential 0 1 4 5 8 9 12 13 16 17 Row/Vertical Odd Field Column/Horizontal a b c d e f g h i j k l mn o p q r 2.2 Programming the display window Five register (DROWS, DROWE, DCOLS, DCOLE and DWLSB) are provided to program the display window as described in the register section of this datasheet. 3.0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 2 3 6 7 10 11 14 15 18 19 Even Field Figure 15: Interlace Read Out Mode Hence, for the example shown in Figure 15, the display window is broken up into two fields, as shown in Figure 15. Pixels a0,b0,...,r0 and a1,b1,...,r1 are readout first and so on until pixels a17,b17,...r17 in the even field are read out. The even field read out is followed by pixels in the odd field, a2,b2,...,r2 then a3,b3,...,r3 until pixels a19,b19,...,r19 11 www.national.com LM9627 Functional Description (continued) 4.0 SUBSAMPLING MODES 4.1 2:1 Sub-Sampling The timing and control circuit can be programmed to sub-sample pixels in the display window vertically, horizontally or both, with an aspect ratio of 2:1 as illustrated in Figure16. 4.2 4:2 Sub-Sampling The timing and control circuit can be programmed to sub-sample pixels in the display window vertically, horizontally or both, with an aspect ratio of 4:2 as illustrated in Figure17 Column/Horizontal a b c d e f g h i j k l mn o p q r 0 1 0 1 2 3 4 5 6 7 8 9 Row/Vertical Row/Vertical Column/Horizontal a b c d e f g h i j k l mn o p q r 2 3 4 5 6 7 8 9 a) Horizontal Sub-sampling a) Horizontal Sub-Sampling Column/Horizontal a b c d e f g h i j k l mn o p q r 0 1 2 3 4 5 6 7 8 9 Row/Vertical Row/Vertical Column/Horizontal a b c d e f g h i j k l mn o p q r 0 1 2 3 4 5 6 7 8 9 b) Vertical Sub-sampling b) Vertical Sub-Sampling Column/Horizontal a b c d e f g h i j k l Column/Horizontal a b c d e f g h i j k l mn o p q r l n o p q r 0 1 2 3 4 5 6 7 8 9 Row/Vertical Row/Vertical LM9627 Functional Description (continued) 0 1 2 3 4 5 6 7 8 9 c) Horizontal & Vertical Sub-sampling c) Horizontal & Vertical Sub-Sampling Green Pixel Green Pixel Red Pixel Blue Pixel Not Read Out Red Pixel Blue Pixel Figure 17: Example 4:2 Sub-sampling Not Read Out Figure 16: Example of 2:1 Sub-sampling Confidential 12 www.national.com 5.0 SNAPSHOT MODE The LM9627 is capable of capturing a single frame of an image under hardware or software control, with or without the aid of an external shutter. Two registers, SNAPSHOTMODE0 & SNAPSHOTMODE1, are provided to program, monitor and control all snapshot sequences. 5.1 Software Controlled Snapshots The snapshot mode events can be software controlled by writing to and reading from the snapshot mode registers over the I2C compatible interface. 5.2 Hardware Controlled Snapshots Two dedicated pins are provided on the LM9627, snapshot & extsync, allowing the snapshot mode events to be controlled by hardware. The snapshot pin must be enabled by writing to the SnapEnable bit of the MCFG0 register. 5.3 Auto Snapshot Mode In auto snapshot mode (see figure 20), upon the receipt of a snapshot or FTriggerNow trigger signal, the integrated timing and control circuit will set the FTriggerEN bit and generate an internal TRIGGER signal (see figure 19), thus resetting the array one row at a time. At end of the reset cycle the timing and control circuit will signal the shutter to open via extsync pin or FtSync bit. At the end of the programmed integration time the shutter will be signalled to close, and the pixel read-out will commence as shown in figure 18a. At the end of the read-out sequence the FTriggerEN will be automatically reset and the sensor will return to video capture mode as shown in figure 20. If an external shutter is not available then at least two frames need to be taken so that the pixels can be integrated over one frame as shown in Figure 18b. To use auto snapshot mode the SsEngage bit of the SNAPSHOTMODE1 register must be set to zero. Array reset, programmable 1 to 4 frames snapshot or FTriggerNow Capture Data image read-out note 1 note 2 irq note 3 FTriggerEn extsync or FtSync FtBusy Start Snapshot Sequence Start of Array Reset Frames Open Shutter Close shutter & start read-out Read-out complete a) With External Shutter Array reset, Capture Data programmable 1 to 4 frames image read-out snapshot or FTriggerNow irq note 1 note 2 note 3 FTriggerEn extsync or FtSync FtBusy Start Snapshot sequence Start of Array Reset Frames Integration Start Start Read-out Bold external pins italic register bits Read-out Complete b) Without External Shutter Note 1: Note 2: Note 3: This wave form shows the snapshot pin programmed to the default pulse mode. The irq pulse is taken low when the snapshot trigger interrupt flag (SsTrigFlag) in the snapshot mode1 (SNAPMODE1) register is read. The irq pulse is taken low when the snapshot trigger interrupt flag (SsRdFlag) in the snapshot mode1 (SNAPMODE1 ) register is read. Figure 18. Snapshot Mode Confidential 13 www.national.com LM9627 Functional Description (continued) snapshot SnapShotPol TRIGGER VIDEO SnapEnable c: SsRdFlag && (SnapshotMod || (SnapshotMod && TRIGGER)) FTriggerNow Figure 19. Snapshot Trigger Generation Logic VIDEO c: SnapshotMod || (SnapshotMod && TRIGGER) LM9627 Functional Description (continued) c:TRIGGER==1 SNAP PREVIEW a:FTriggerEn=1 c:TRIGGER==1 IRQ a:SsTrigFlag=1 c:FTriggerEn==1 SNAP a:FTriggerEn=0 PREVIEW a: SsRdFlag = 1 a: FtTriggerEn = 0 Figure 20. Auto Snapshot Mode State Diagram 5.4 CPU Snapshot Mode In CPU snapshot mode, the FTriggerEN is not set automatically and an Interrupt generator can be enabled. Hence, upon the receipt of a snapshot or FTriggerNow trigger signal, the integrated timing and control circuit will generate an internal TRIGGER signal as shown in figure 19 and then wait in the IRQ state for the FTriggerEN bit to be manually set as shown in figure 21. Once the FtriggerEn bit is set the integrated timing and control circuit will start resetting the array one row at a time. At end of the reset cycle the timing and control circuit will signal the shutter to open via extsync pin or FtSync bit. At the end of the programmed integration time the shutter will be signalled to close, and the pixel read-out will commence as shown in figure 18a. At the end of the read-out sequence the FTriggerEN will be automatically disabled and the sensor will return to video capture mode as shown in figure 20. If an external shutter is not available then at least two frames need to be taken so that the pixels can be integrated over one frame as shown in Figure 18b. To use CPU snapshot mode the SsEngage bit of the SNAPSHOTMODE1 register must be set to one. An interrupt generator can be enabled in CPU snapshot mode by setting the SnapIntEn bit of SNAPSHOTMODE1 register. An interrupt will be generated on the external interrupt pin, irq, when a snapshot sequence is triggered (TRIGGER=1) or when the array readout is complete at the end of the snapshot sequence as shown figure 21. Confidential Figure 21. CPU Snapshot Mode State Diagram When an interrupt is generated by a TRIGGER event, the SsTrigFlag bit in the SNAPSHOTMODE1 register is set. Similarly when an interrupt is generated at the completion of a readout the SsRdFlag in the SNAPSHOTMODE1 register is set. The polarity of the irq pin can be programmed. The interrupt can only be cleared by reading SsTrigFlag and the SsRdFlag as shown in figure 22. SsTrigFlag SsRdFlag irq SnapIntEn IrqPol Figure 22. Interrupt Request Generation Logic 5.5 Pulse & Level Trigger Mode The snapshot pin can be programmed to operate in pulse trigger mode where one snapshot sequence is executed per active pulse or in level trigger mode where by snapshot sequences are repeated as long as the level on the snapshot pin is held active. (see figures 20 and 21). Pulse and level trigger modes can be set by programming the SnapshotMod bit in the SNAPSHOTMODE0 register. 14 www.national.com 6.0 CLOCK GENERATION MODULE The LM9627 contains a clock generation module that will create two clocks as follows: Hclk, the horizontal clock. This is an internal system clock and can be programmed to be the input clock (mclk) or mclk divided by any number between 1 and 255. CLKpixel the pixel clock. This is the external pixel clock that appears at the digital video port. It can be Hclk or Hclk divided by 2. This clock cannot be programed. 7.0 The number of rows in a scan window is given by: SWN rows = (RADend - RADstart) + 1 Where: RADend RADstart is the end row address of the defined scan window. (See section 2.1) is the start row address of the defined scan window. (Scan section 2.1). The number of Hclk clocks required to process a full frame is given by: FRAME RATE PROGRAMING A frame is defined as the time it takes to reset every pixel in the array, integrate the incident light, convert it to digital data and present it on the digital video port. This is not a concurrent process and is characterized in a series of events each needing a certain amount of time as shown in Figure 23. FN Hclk = [(Mfactor * SWN rows ) + Fdelay ] * RNHclk Where: Mfactor Start Row address = 0 Progressive Scan 1 Sub-sampling or Interlace 0.5 SWN rows is the Number of Rows in Selected Scan Window. Fdelay a programmable value between 0 & 4097 repre- Row delay time senting the Inter Frame Delay in multiples of RN Hclk . This parameter allows the frame time to be extended. (See the Frame Delay High and Frame Delay Low registers). Row Time Transfer all pixels to CDS Reset all pixels in row is a Mode Factor which must be applied. It is dependent on the selected mode of operation as shown in the table below: The frame rate is given by: Frame Rate = Shift all pixels out of row 7.2 Partial Frame Integration In some cases it is desirable to reduce the time during which the pixels in the array are allowed to integrate incident light without changing the frame rate. Row address + 1 Yes Last row? No Figure 23. Frame Readout Flow Diagram 7.1 Full Frame Integration Full frame integration is when each pixel in the array integrates light incident on it for the duration of a frame (see Figure 24). The number of Hclk clock cycles required to process & shift out one row of pixels is given by: RN Hclk = R opcycle + R delay Where: R opcycle R delay Confidential Hclk FN Hclk This is known as Partial Fame Integration and can be achieved by resetting pixels in a given row ahead of the row being selected for readout as shown in Figure 24. The number of Hclk clocks required to process a partial frame is given by: FPHclk = RN Hclk * Itime Where: RN Hclk Itime is the number of Hclk clock cycles required to process & shift out one row of pixels. is the number of rows ahead of the current row to be reset. (See the Integration Time High and Low registers). The Integration time is subject to the following limits: is a fixed integer value of 780 representing the Row Operation Cycle Time in multiples of Hclk clock cycles. It is the time required to carry out all fixed row operations outlined in Figure 23. a programmable value between 0 & 2047 representing the Row Delay Time in multiples of Hclk. This parameter allows the Row Operation Cycle time to be extended. (See the Row Delay High and Row Delay Low registers). Mode 15 Limit Progressive Scan Itime <= SWNrows + Fdelay Interlace Itime <= SWNrows + 2* Fdelay Sub-Sampled Itime <= SWNrows + 0.5*Fdelay www.national.com LM9627 Functional Description (continued) LM9627 Functional Description (continued) Full Integration Time Partial Integration Time Row n Frame Delay Row 0 Row 1 Row 2 Row x Programmable Row Delay Row x+ Row n Frame Delay Row 0 Row CDS, Reset Row x & Shift Full Frame integration Programmable Row Delay Row CDS, Reset Row x+ & Shift Partial Frame Integration Frame N Figure 24. Partial and Full Frame Integration 7.3 Frame Rate Programming Guide The table bellow can be used as a guide for programming the sensor. Note that it is assumed that the sensor is being driven with a 48MHz clock. All programmed values are given in decimal. register vclkgen rdelayh rdelayl fdelayh fdelayl srows srowe dwlsb address 05hex 15hex 16hex 17hex 18hex 0Bhex 0Chex 12hex [10:8] [7:0] [11:8] [7:0] [8:1] [8:1] fps 30 4 0 0 0 9 0 251 50 15 4 0 0 2 40 0 251 50 7.5 4 0 0 6 12 0 251 50 3.75 4 3 12 6 12 0 251 50 25 4 0 172 0 0 0 251 50 12.5 5 0 0 1 226 0 251 50 6.25 5 0 0 5 188 0 251 50 3.125 4 0 156 14 14 0 251 50 5 4 2 255 4 23 0 251 50 4 5 0 0 10 12 0 251 50 3 5 0 0 14 14 0 251 50 2 6 0 200 13 248 0 251 50 1 6 3 241 15 126 0 251 50 Confidential 16 www.national.com 10.0 ANALOG GAIN ADJUSTMENT 8.0 The integrated analog programmable gain amplifier is capable of applying a linear gain 1X to 5.6X in 64 linear steps. This can be programmed using the VGAIN register as shown in the table SIGNAL PROCESSING 8.1 Bad Pixel Detection & Correction The LM9627 has a built-in bad pixel detection and correction block that operates on the fly. This block can be switched off by the user. below: 8.2 Black Level Compensation In addition to the programmable gain the LM9627 has a built in black level compensation block as illustrated in Figure 25. This block can be switched off. *a *(1-a) + + z-1 + compensated output only enabled for black pixels input signal Figure 25. Digital Black Level Compensation. The black level compensation block will subtract the average signal level of the black pixels around the array from the digital video output to compensate for the temperature and integration time dependent dark signal level of the pixels. The exponential averaging circuit shown in figure 25 only operates on the least significant 8 bits of the video data. 9.0 POWER MANAGMENT 9.1 Power Up and Down The LM9627 is equipped with an on-board power management system allowing the analog and digital circuitry to be switched off (power down) and on (power up) at any time. The sensor can be put into power down mode by asserting a logic one on the "pdwn" pin or by writing to the power down bit in the main configuration register via the I2 C compatible serial interface. To power up the sensor a logic zero can be asserted on the "pdwn" pin or write to the power down bit in the main configuration register via the I2 C compatible serial interface. It will take a few milli seconds for all the circuits to power up. The power management register contains a bit indicating when the sensor is ready for use. During this time the sensor cannot be used for capturing images. A status bit in the power management register will indicate when the sensor is ready for use. 9.2 Advanced Power Features In addition to the power up/power down features of the sensor, sections of the analog video processing chain can be powered down and re-routed during normal operation. This flexibility allows power dissipation to be traded of with signal gain as shown in the table below: PGA Amp Power Saving on 0mW off 10mW VidGain VidGain VidGain Hex Code Gain Amp Value VidGain Dec Code Dec Code Hex Code Gain Amp Value 0 00 1 32 20 3.34 1 01 1.07 33 21 3.41 2 02 1.15 34 22 3.48 3 03 1.22 35 23 3.56 4 04 1.29 36 24 3.63 5 05 1.37 37 25 3.7 6 06 1.44 38 26 3.77 7 07 1.51 39 27 3.85 8 08 1.58 40 28 3.92 9 09 1.66 41 29 3.99 10 0A 1.73 42 2A 4.07 11 0B 1.8 43 2B 4.14 12 0C 1.88 44 2C 4.21 13 0D 1.95 45 2D 4.29 14 0E 2.02 46 2E 4.36 15 0F 2.1 47 2F 4.43 16 10 2.17 48 30 4.5 17 11 2.24 49 31 4.58 18 12 2.31 50 32 4.65 19 13 2.39 51 33 4.72 20 14 2.46 52 34 4.8 21 15 2.53 53 35 4.87 22 16 2.61 54 36 4.94 23 17 2.68 55 37 5.02 24 18 2.75 56 38 5.09 25 19 2.83 57 39 5.16 26 1A 2.9 58 3A 5.23 27 1B 2.97 59 3B 5.31 28 1C 3.04 60 3C 5.38 29 1D 3.12 61 3D 5.45 30 1E 3.19 62 3E 5.53 31 1F 3.26 63 3F 5.6 Figure 26. Power Control Confidential 17 www.national.com LM9627 Functional Description (continued) LM9627 Functional Description (continued) 11.0 OFFSET ADJUSTMENT 12.0 For maximum image quality over a wide range of light conditions it is necessary to set an appropriate offset voltage before using the sensor to capture images. This offset voltage must be applied to the offset pin (38) of the sensor, and is used to adjust the analogue video signal being fed to the internal A/D. The fine offset adjustment and calibration method described in section 11.0 will ensure that the sensor's black level is optimized for a fixed analog gain setting. However, when the analog gain is changed substantially, the black level of the sensor will shift resulting in a white washed image. To stop this effect from occurring, the black level needs to be recalibrated. This can be done as part of the contrast adjustment which is carried out by most digital image processors. If this is not possible then the following method can be used. The relationship between the gain and the offset can be described with the following equation. The level of the offset voltage determines the black level of the image and has a direct impact on the image quality. Too high an offset results in a white washed or hazy looking image, while too low of an offset results in a dark image with low contrast even though the light conditions are good. Offset(G) = Offset(0) + C * G0.4 A fine offset adjustment should be applied to each part by programming the offset voltage via the I2 C compatible serial interface. To program an offset voltage the following procedure should be followed: The sensor's offset, fine_i & fine_ctrl pins should be connected as shown in figure 2. The following procedure should be followed to calibrate the offset * Disable the black level compensation block by writing a logic 1 to bit 4 of the Main Configuration Register 0 (MCFG0: address 02Hex). * The offset can be adjusted by writing to the Offset Compensation Registers (OCR: addresses 1F, 22 & 25 hex). Writing 00hex will give the largest voltage, while writing FF hex will give the smallest value. * Run the following binary search algorithm * For n=7 to 0 step -1 * { Set bit n in the OCR registers (addresses 1F, 22 & 25 Hex) to a logic one by writing over the I 2 C compatible interface. Read a full frame and calculate the average black level (BLaverage) of the first and last 5 black pixels in the every row of the array If (BLaverage < 100) then Reset bit n in the OCR registers (addresses 1F, 22 & 25 Hex) to 0 else Keep bit n set to one. } * Enable the black level compensation block (if desired) by writing a logic 0 to bit 4 of the Main Configuration Register 0 (MCFG0: address 02Hex). Confidential OFFSET & GAIN where: Offset(G) is the offset that needs to be programmed in the OCR1, OCR2 & OCR3 registers to ensure the correct black level setting for an analog gain setting of G. Offset(0) is the offset that needs to be programmed in the OCR1, OCR2 & OCR3 registers to ensure the correct black level setting for unity analog gain, (G=0). C is a constant and will vary from sensor to sensor G is the value programmed in the VGAIN register of the sensor which determines the sensor's analog gain. The following procedure should be used to calculate the value of C: Use the calibration procedure described in section 11.0 to determine the offset at unity gain, offset(0). Note the VGAIN register should be set to 0. Set the sensor's analog gain register (VGAIN) to its max setting, 31, and repeat the calibration procedure described in section 11.0. This will allow the offset at full gain, 31, that needs to be programmed in the OCR1, OCR2 & OCR3 registers to ensure the correct black level setting to be determined. The value of C for a particular sensor can be calculated using the following formula: C= Offset(31) - Offset(0) 3.95 Once the value of C has been calculated, offset values for different gain settings can be calculated using equation 1. It is recommended that a two decimal point accuracy for C is maintained. 18 www.national.com 13.0 SERIAL BUS The serial bus interface consists of the sda (serial data), sclk (serial clock) and sadr (device address select) pins. The LM9627 can operate only as a slave. 13.4 Data Valid The master must ensure that data is stable during the logic 1 state of the sclk pin. All transitions on the sda pin can only occur when the logic level on the sclk pin is "0" as shown in Figure 29. The sclk pin is an input, it only and controls the serial interface, all other clock functions within LM9627 use the master clock pin, mclk. 13.1 Start/Stop Conditions The serial bus will recognize a logic 1 to logic 0 transition on the sda pin while the sclk pin is at logic 1 as the start condition. A logic 0 to logic 1 transition on the sda pin while the sclk pin is at logic 1 is interrupted as the stop condition as shown in Figure 27. sda sclk S P start condition stop condition Figure 27. Start/Stop Conditions 13.2 Device Address The serial bus Device Address of the LM9627 is set to 1010101 when sadr is tied low and 0110011 when sadr is tied high. The value for sadr is set at power up. 13.3 Acknowledgment The LM9627 will hold the value of the sda pin to a logic 0 during the logic 1 state of the Acknowledge clock pulse on sclk as shown in Figure 28. sda from master MSB ACK sda from sensor sclk ACK S 1 7 2 8 9 Clock pulse for ACK START Figure 28. Acknowledge sda sclk data line stable; data valid change of data allowed data line stable; data valid Figure 29. Data Validity 13.5 Byte Format Every byte consists of 8 bits. Each byte transferred on the bus must be followed by an Acknowledge. The most significant bit of the byte is should always be transmitted first. See Figure 30. 13.6 Write Operation A write operation is initiated by the master with a Start Condition followed by the sensor's Device Address and Write bit. When the master receives an Acknowledge from the sensor it can transmit 8 bit internal register address. The sensor will respond with a second Acknowledge signaling the master to transmit 8 write data bits. A third Acknowledge is issued by the sensor when the data has been successfully received. The write operation is completed when the master asserts a Stop Condition or a second Start Condition. See Figure 31. 13.7 Read Operation A read operation is initiated by the master with a Start Condition followed by the sensor's Device Address and Write bit. When the master receives an Acknowledge from the sensor it can transmit the internal Register Address byte. The sensor will respond with a second Acknowledge. The master must then issue a new Start Condition followed by the sensor's Device Address and read bit. The sensor will respond with an Acknowledged followed by the Read Data byte. The read operation is completed when the master asserts a Not Acknowledge followed by Stop Condition or a second Start Condition. See Figure 32. MSB sda ack signal from receiver ack signal from receiver byte complete sclk 1 7 2 9 8 1 2 ACK clock line 8 9 ACK held low S START P Figure 30. Serial Bus Byte Format Device Address S W A Register Address A Data Byte A P bold sensor action Figure 31. Serial Bus Write Operation S Device Address W A Register Address A S Device Address R A Data Byte _ A P bold sensor action Figure 32. Serial Bus Read Operation Confidential 19 www.national.com LM9627 Functional Description (continued) LM9627 Functional Description (continued) 14.0 DIGITAL VIDEO PORT The captured image is placed onto a flexible 12-bit digital port as shown in Figure 10. The digital video port consists of a programmable 12-bit digital Data Out Bus (d[11:0]) and three programmable synchronisation signals (hsync, vsync, pclk). This feature allows a programmable digital gain to be implemented when connecting the sensor to 8 or 10 bit digital video processing systems as illustrated in Figure 34. The unused bits on the digital video bus can be optionally tri-stated. By default the synchronisation signals are configured to operate in "master" mode. They can be programed to operate in "slave" mode. LM9627 The following sections are a detailed description of the timing and programming modes of digital video port. Pixel data is output on a 12-bit digital video bus. This bus can be tri-stated by asserting the TriState bit in the VIDEOMODE1 register. d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 10 bit Digital Image Processor a) LM9627 Connected to a 10 bit Digital Image Processors 14.1 Digital Video Data Out Bus (d[11:0]) A programmable matrix switch is provided to map the output of the internal pixel framer to the pins of the digital video bus as illustrated in Figure 33. Internal Pixel Framer Output Register 11 10 9 8 7 6 5 4 3 2 1 0 LM9627 d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 a) MSB Bit 11, Switch Mode (default) d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 d7 d6 d5 d4 d3 d2 d1 d0 8 bit Digital Image Processor b) LM9627 Connected to a 8 bit Digital Image Processors Figure 34. Example of connection to 10/8 bit systems Internal Pixel Framer Output Register 11 10 9 8 7 6 5 4 3 2 1 0 d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 b) MSB Bit 10, Switch Mode Internal Pixel Framer Output Register 9 11 10 8 7 6 5 4 3 2 1 0 Synchronisation Signals in Master Mode By default the sensor's digital video port's synchronisation signals are configured to operate in master mode. In master mode the integrated timing and control block controls the flow of data onto the 12-bit digital port, three synchronisation outputs are provided: pclk is the pixel clock output pin. hsync is the horizontal synchronisation output signal. vsync is the vertical synchronisation output signal. 14.2 Pixel Clock Output Pin (pclk) (Master Mode) The pixel clock output pin, pclk, is provided to act as a synchronisation reference for the pixel data appearing at the digital video out bus pins d[11:0]. This pin can be programmed to operate in two modes: * In free running mode the pixel clock output pin, pclk, is always running with a fixed period. Pixel data appearing on the digital video bus d[11:0] are synchronized to a specified active edge of the clock as shown in Figure 35. pclk d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 d[11:0] c) MSB bit 9, Switch Mode a) pclk active edge negative Internal Pixel Framer Output Register 11 10 9 8 7 6 5 4 3 2 1 0 pclk d[11:0] d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 d) MSB bit 8, Switch Mode Figure 33. Digital Video Bus Switching Modes Confidential b) pclk active edge positive (default) invalid pixel data Figure 35. pclk in Free Running Mode * In data ready mode, the pixel clock output pin (pclk) will produce a pulse with a specified level every time valid pixel data appears on the digital video bus d[11:0] as shown in Figure 36. 20 www.national.com , 14.4 Vertical/Horizontal Synchronisation Pin (vsync) The vertical synchronisation output pin, vsync, is used as an indicator for pixel data within a frame. The vsync output pin can be programmed to operate in two modes as follows: pclk d[11:0] a) pclk active edge negative pclk d[11:0] b) pclk active edge positive invalid pixel data Figure 36. pclk in Data Ready Mode By default the pixel clock is a free running active low (pixel data changes on the positive edge of the clock) with a period equal to the internal hclk. The active edge of the clock can be programmed such that pixel data changes on the positive or negative edge of the clock. * Level mode should be used when the pixel clock, pclk, is programmed to operate in free running mode. In level mode the vsync output pin will go to the specified level (high or low) at the start of each frame and remain at that level until the last pixel of that row in the frame is placed on d[11:0] as shown in Figure 39. The hsync level is always synchronized to the active edge of pclk. pclk d[11:0] vsync Frame n+1 Frame n a) vsync programmed to be active high 14.3 Horizontal Synchronisation Output Pin (hsync) The horizontal synchronisation output pin, hsync, is used as an indicator for row data. The hsync output pin can be programmed to operate in two modes as follows: * Level mode should be used when the pixel clock, pclk, is programmed to operate in free running mode. In level mode the hsync output pin will go to the specified level (high or low) at the start of each row and remain at that level until the last pixel of that row is read out on d[11:0] as shown in Figure 37. The hsync level is always synchronized to the active edge of pclk. pclk d[11:0] hsync Row n Row n+1 a) hsync programmed to be active high (default) pclk d[11:0] vsync pclk d[11:0] d[11:0] hsync pclk d[11:0] hsync Row n+1 Row n a) hsync programmed to be active high vsync d[11:0] vsync Frame n Frame n+1 b) vsync programmed to be active low (default) invalid pixel data Figure 40. vsync in pulse mode 14.5 Odd/Even Mode In odd/even mode the vsync signal is used to indicate when pixel data from an odd and even field is being placed on the digital video bus d[11:0]. The polarity of vsync can still be programmed in this mode as shown in Figure 41 pclk d[11:0] d[11:0] vsync Row n Row n+1 Odd Field Even Field a) vsync programmed to be active high (default) b) hsync programmed to be active low invalid pixel data Figure 38. hsync in Pulse Mode By default the first pixel data at the beginning of each row is placed on the digital video bus as soon as hsync is activated. It is possible to program up to 15 dummy pixels to be readout at the beginning of each row before the real pixel data is readout. This feature is supported for both level and pulse mode. Confidential Frame n+1 Frame n a) vsync programmed to be active high pclk pclk hsync Frame n+1 * Pulse mode should be used when the pixel clock, pclk, is programmed to operate in data ready mode. In pulse mode the vsync output pin will produce a pulse at the end of each frame. The width of the pulse will be a minimum of four hclk cycles and its polarity can be programmed as shown in Figure 40. The vsync level is always synchronized to the active edge of pclk. pclk Row n+1 Row n b) hsync programmed to be active low invalid pixel data Figure 37. hsync in Level Mode * Pulse mode should be used when the pixel clock, pclk, is programmed to operate in data ready mode. In pulse mode the hsync output pin will produce a pulse at the end of each row. The width of the pulse will be a minimum of four pclk cycles and its polarity can be programmed as shown in Figure 38. The hsync level is always synchronized to the active edge of pclk Frame n b) vsync programmed to be active low invalid pixel data Figure 39. vsync in Level Mode pclk d[11:0] vsync Odd Field Even Field b) vsync programmed to be active low invalid pixel data Figure 41. vsync in odd/even Mode 21 www.national.com LM9627 Functional Description (continued) LM9627 Functional Description (continued) pclk vsync hsync c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 d[11:0] c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 row 2 row1 c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 row 1 row 2 frame 1 frame 2 Programmable hsync to 1st valid pixel delay Programmable inter-frame delay Programmable row delay Figure 42. Example of Digital Video Port Timing in Progressive Scan Mode pclk vsync hsync c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 d[11:0] c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 row1 c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 row 3 c0 c1 c2 c3 c4 c5 c6 c7 c8 c9 row 2 row 4 Odd Field Even Field Programmable hsync to 1st valid pixel delay Programmable inter-frame delay Programmable row delay Figure 43. Example of Digital Video Port Timing in Interlaced Mode pclk vsync hysync c0 d[11:0] c2 c4 c6 c8 c0 row 1 c2 c4 c6 c8 c0 row 3 c2 c4 c6 c8 c0 c2 c4 row 1 c6 c8 row 3 frame 1 frame 2 Programmable hsync to 1st valid pixel delay Programmable inter-frame delay Programmable inter-row delay Figure 44. Example of Digital Video Port Timing in 2:1 Sub-sampling Mode pclk vsync hsync c0 d[11:0] c2 c4 c6 c8 c0 row 1 c2 c4 c6 c8 c0 row 2 c2 c4 c6 c8 c0 row 1 frame 1 c2 c4 c5 c8 row 2 frame 2 Programmable hsync to 1st valid pixel delay Programmable inter-frame delay Programmable inter-row delay Figure 45. Example of Digital Video Port Timing in 4:2 Sub-sampling Mode Confidential 22 www.national.com 14.6 Synchronisation Signals in Slave Mode The sensor's digital video port's synchronisation signals can be programmed to operate in slave mode. In slave mode the integrated timing and control block will only start frame and row processing upon the receipt of triggers from an external source. Only two synchronization signals are used in slave mode as follows: hsync is the row trigger input signal. vsync is the frame trigger input signal. Figure 46 shows the LM9627's digital video port in slave mode connected to a digital video processor master DVP. d[11:0] din[11:0] hsync RowTrig vsync FrameTrig Xmclk = 124 + DWStAd Where: DW StAd is the value of the display window column start address. The polarity of the active level of the row trigger is programmable. By default it is active high. 14.8 Frame Trigger Input Pin (vsync) The frame trigger input pin, vsync, is used to reset the row address counter and prepare the array for row processing. It must be activated for at least one "mclk" cycle and no more than 96 mclk cycles after the activation of hsync as illustrated in Figure 48. pclk mclk 14.7 Row Trigger Input Pin (hsync) The row trigger input pin, hsync, is used to trigger the processing of a given row. It must be activated for at least two "mclk" cycle. The first pixel data will appear at d[11:0] "Xmclk "periods after the assertion of the row trigger, were Xmclk is given by: MasterClock LM9627 DVP Figure 46. LM9627 in slave mode The polarity of the active level of the row trigger is programmable. By default it is active high. 780 clock cycles per line hsync pixel 12 pixel 11 pixel 652 d[11:0] 642 valid pixels mclk count 776 777 778 779 0 1 2 3 ... 134 135 136 136 137 ... 774 775 776 777 778 779 0 1 mclk Figure 47. hsync slave mode timing diagram for centred display window of 642 pixels 780 clock cycles per line hsync No more than 96 clock cycles vsync internal row counter mclk count line502 line 502 776 777 778 779 0 1 2 3 ... line503 774 775 776 777 778 779 0 1 ... line 0 774 775 776 777 778 779 0 1 mclk Figure 48. vsync slave mode timing diagram for scan window of 504 rows. Confidential 23 www.national.com LM9627 Functional Description (continued) LM9627 MEMORY MAP ADDR Register Reset Value 00h Reserved for future use. 01h REV 02h Revision Register 02h MCFG0 00h Main Configuration Register 0 03h MCFG1 00h Main Configuration Register 1 04h PCR 00h Power Control Register. 05h VCLKGEN 04h Video Clock Generator 06h VMODE0 00h Video Mode 0 Register 07h VMODE1 00h Video Mode 1 Register 08h VMODE2 00h Video Mode 2 Register 09h SNAPMODE0 00h Snapshot Mode 0 Register 0Ah SNAPMODE1 00h Snapshot Mode 1 Register 0Bh SROWS 00h Scan Window Row Start Register 0Ch SROWE FBh Scan Window Row End Register 0Dh Confidential Description Reserved for future use. 0Eh DROWS 00h Display Window Row Start Register 0Fh DROWE FBh Display Window Row End Register 10h DCOLS 00h Display Window Column Start Register 11h DCOLE A5h Display Window Column End Register 12h DWLSB 32h Display Window LSB Register. 13h ITIMEH 00h Integration Time High Register 14h ITIMEL 00h Integration Time Low Register 15h RDELAYH 00h Row Delay High Register 16h RDELAYL 00h Row Delay Low Register 17h FDELAYH 00h Frame Delay High Register 18h FDELAYL 00h Frame Delay Low Register 19h VGAIN 00h Video Gain Register 1Fh OCR1 00h Offset Compensation Register 1 22h OCR1 00h Offset Compensation Register 1 25h OCR2 00h Offset Compensation Register 2 26h BLCOEFF 00h Black Level Compensation Coefficient Register 27h BPTH0H 00h Bad pixel Threshold 0 High Register 28h BPTH0L 00h Bad pixel Threshold 0 Low Register 29h BPTH1H 00h Bad pixel Threshold 1 High Register 2Ah BPTH1L 00h Bad pixel Threshold 1 Low Register 24 www.national.com LM9627 Register Set The following section describes all available registers in the LM9627 register bank and their function. Register Name Mnemonic Address Type Bit 7:0 Bit Symbol SiRev Register Name Address Mnemonic Type: Reset Value Bit 7 6 5 Device Rev Register REV 01 Hex Read Only. Bit Description Main Configuration 1 03 Hex MCFG1 Read/Write 00 Hex Bit Symbol ColorMode Assert when using a monochrome sensor. When this bit is at a logic 1, Sub-Sampling is set to 2:1 and every other row is read out during interlace mode. Clear (the default) when using a color sensor. When this bit is at logic 0, sub-sampling is set to 4:2 and every other row pair is read out during interlace mode. 6 ScanMode Assert to set the sensor to interlace readout mode. Clear (the default) to set the sensor to progressive scan read out mode. 5 HSubSamEn Assert to enable horizontal subsampling. Clear (the default) to disable horizontal sub-sampling. 4 VSubSamEn Assert to enable vertical subsampling. Clear (the default) to disable vertical sub-sampling. Main Configuration 0 02 Hex MCFG0 Read/Write 00 Hex PwrUpBusy PwrDown BPCorrection Description (Read Only Bit) Indicates that power on initialization is in progress. The sensor is ready for use when this bit is at logic 0. Assert to power down the sensor. Writing a logic 1 to this register bit has the same effect as taking the pdwn pin high. Clear (the default) this bit to power up the sensor. Assert to enable the bad pixel detection and correction circuit. Clear (the default) to switch it off. 4 BlkLComp Assert to disable the black level compensation circuit. Clear (the default) to switch it on. 3 SnapEnable Assert to enable the external snapshot pin. Clear (the default) to disable the external snapshot pin. Reserved 3 2 Reserved SlaveMode 1:0 Use to configure the digital video port's synchronisation signal to operate in slave mode. By default the digital video's port's synchronization signals are configured to operate in master mode. Reserved Register Name Address Mnemonic Type Power Control Register 1 04 Hex PCR Read/Write Reset Value 00 Hex Bit 7 Bit Symbol ByPassGain 6:4 3 0 25 Description Assert to route the analog video signal from the output of the CDS to the input of the 12 bit A/D. Clear (the default) to route the signal to the video gain amplifier. Reserved PwdnPGA 2:1 Confidential Description 7 The silicon revision register. Bit Symbol 2:0 Register Name Address Mnemonic Type Reset Value Assert to power down the programmable video gain amplifier. Clear (the default) to power up the video gain amplifiers. Reserved PwDnADC Assert to power down the 12 bit analog to digital convertor. Clear (the default) to power up the 12 bit analog to digital convertor. www.national.com LM9627 Register Set (continued) Register Name Address Mnemonic Type Reset Value Bit 7:0 Bit Symbol HclkGen Register Name Address Mnemonic Type Reset Value Bit 7:6 Hclk Generator Register 05 Hex VCLKGEN Read/Write 04 Hex. Use to divide the frequency of the sensors master clock input, mclk to generate the internal sensor clock, Hclk. Program 00 Hex (the default) for Hclk to equal mclk or divide mclk by any number between 1 and FF Hex. Digital Video Mode 0 06 Hex VMODE0 Read/Write 00 Hex Bit Symbol PixDataSel PixDataMsb 3:0 7 PixClkMode 6 VsyncMode 5 HsyncMode 4 PixClkPol 3 VsynPol 2 HsynPol 1 OddEvenEn 0 TriState Description Description Use to program the number of active bits on the digital video bus d[11:0], starting from the MSB (d[11]). Inactive bits are tri-stated.: 00 5:4 Register Name Digital Video Mode 1 Address 07 Hex Mnemonic VMODE1 Type Read/Write Reset Value 00 Hext Bit Bit Symbol Description 12 bit mode, bits d[11:0] of the digital video bus are active. This is the default. 01 10 bit mode, bits d[11:2] of the digital video bus are active. 10 8 bit mode, bits d[11:4] of the digital video bus are active. 11 Reserved. Use to program the routing of the MSB output of the internal video A/D to a bit on the digital video bus. 00 A/D [11:0] -> d[11:0]. 01 A/D [10:0] -> d[11:1] 10 A/D [9:0] -> d[11:2] 11 A/D [8:0] -> d[11:3] Reserved Register Name Address Mnemonic Type Reset Value Bit 7:4 3:0 Confidential 26 Assert to set the pclk to "data ready mode". Clear, the default, to set pclk to "free running mode". Assert to set the vsync pin to "pulse mode". Clear (the default) to set the vsync signal to "level mode". Assert to force the hsync signal to pulse for a minimum of four pixel clocks at the end of each row. Clear (the default) to force the hsync signal to a level indicating valid data within a row. Assert to set the active edge of the pixel clock to negative. Clear (the default) to set the active edge of the clock to positive. Assert to force the vsync signal to generate a logic 0 during a frame readout (Level Mode), or a negative pulse at the end of a frame readout (Pulse Mode). Clear (the default) to force the vsync signal to generate a logic 1 during a frame readout (Level Mode), or a negative pulse at the end of a frame readout (Pulse Mode). Assert to force the hsync signal to generate a logic 0 during a row readout (Level Mode), or a negative pulse at the end of a row readout (Pulse Mode). Clear (the default) to force the hsync signal to generate a logic 1 during a row readout (Level Mode), or a negative pulse at the end of a readout (Pulse Mode). Assert to force the vsync pin to act as an odd/even field indicator. Clear (the default) to force the vsync pin to act as a vertical synchronization signal. Assert to tri-state all output signals (data and control) on the digital video port. Clear (default) to enable all signals (data and control) on the digital video port. Digital Video Mode 2 08 Hex VMODE2 Read/Write 00 Hex Bit Symbol Description HsyncAdjust Use to program the leading edge of hsync to the first valid pixel at the beginning of each row. This can be 0-hex to F-hex corresponding to 0 - 15 pixel clocks. Default 0. Reserved www.national.com Register Name Address Mnemonic Type Reset Value Bit 7.6 Snapshot Mode Configuration Register 0 09 Hex SNAPMODE0 Read/Write 00 Hex Bit Symbol SsFrames Program to set the number of frames required before readout during a snapshot with no external shutter, (see Figure 18). By default these two bits are set to 00 resulting in one frame before readout: 0 one frame 01 two frames 10 three frames 11 four frames ShutterEn Assert to indicate that an external shutter will be used during snapshot mode. Clear (the default) to indicate that snapshot mode will be carried out without the aid of an external shutter. 4 ExtSynPol Assert to set the active level of the extsync signal to 0. Clear (the default) to set the active level of the extsync signal to 1. 2 1 0 Bit Description 5 3 Register Name Address Mnemonic Type Reset Value Snapshot Mode Configuration Register 1 0A Hex SNAPMODE1 Read/Write 00 Hex. Bit Symbol 7 SnapIntEn Assert to enable the snapshot interrupt generator. Clear (the default) to disable the interrupt generator. 6 SsTrigFlag (Read Only Bit) Snapshot trigger interrupt flag. A logic 1 in this bit indicates that the generated interrupt on the irq pin is due to a snapshot trigger. This bit is cleared when read. 5 SsRdFlag (Read Only Bit) Snapshot read done interrupt flag. A logic 1 in this bit indicates that the generated interrupt on the irq pin is due to the completion of a snapshot readout sequence. This bit is cleared when read. 4 SsEngage Assert to allow a CPU controlled snapshot sequence. In this mode the snapshot trigger will only generate an interrupt to the CPU and the CPU must manually start the snapshot sequence by asserting the FTriggerEn bit of this register. Clear (the default) engage an automatic snapshot sequence. In auto mode the snapshot sequence is started as soon as a snapshot trigger is asserted. 3 FtSync (Read Only Bit) The internal synchronisation signal. A logic 1 on this bit indicates a synchronization event is required. This bit is functionally equivalent to the external extsync pin. 2 FtBusy (Read Only Bit) The Frame Trigger Busy bit. A logic 1 on this bit indicates that the sensor is busy reading out pixel data as shown in Figure 18. 1 FTriggerNow Assert to start a snapshot sequence. The frame trigger now is functionally equivalent to the external snapshot pin. The default is 0. 0 FTriggerEn Assert to enable a snapshot sequence (see the SsEngage bit of this register). The default is 0. Reserved SnapshotMod SnapShotPol IrqPol Confidential Assert to set the snapshot pin to level mode. In level mode the sensor will continually run snapshot sequences as long as the snapshot pin is held to the active level. Clear (the default) to set the snapshot signal to pulse mode. In pulse mode the sensor will only carry out one snapshot sequence per pulse applied to the snapshot pin. Assert to set the snapshot pin to be active on the positive edge. Clear (the default) to set the snapshot pin to be active on the negative edge. Assert to set the active level of the irq signal to 0, Clear (the default) to set the active level of the irq signal to 1. 27 Description www.national.com LM9627 Register Set (continued) LM9627 Register Set (continued) Register Name Address Mnemonic Type Reset Value Bit 7:0 Bit Symbol Description SwStartRow Use to program the scan window's start row address MSBs. If bit 6 of register DWLSB is set to 1 the start row address is incremented by 1 else the raw value is used. Register Name Address Mnemonic Type Reset Value Bit 7:0 Scan Window Row Start Register 0B Hex SROWS Read/Write 00 Hex Scan Window Row End Register 0C Hex SROWE Read/Write FB Hex Bit Symbol SwEndRow Use to program the scan window's end row address MSBs. If bit 6 of register DWLSB is set to 1 the end row address is incremented by 1. else the raw value is used. Display Window Row Start Register 0E Hex DROWS Read/Write Reset Value 00 Hex 7:0 Bit 7:0 Description DwStartRow Use to program the display window's start row address MSBs. The LSB can be programmed using the DWLSB register. Bit 7:0 Description DwEndRow Use to program the scan window's end row address. The LSB can be programmed using the DWLSB register. Display Window Column Start Register 10 Hex DCOLS Read/Write 00 Hex Bit Symbol DwStartCol Register Name Address Mnemonic Type Reset Value Use to program the display window's start column address MSBs. The two LSBs can be programmed using the DWLSB register. Display Window Column End Register 11 Hex DCOLE Read/Write A5 Hex DwEndCol Description Use to program the scan window's end column address MSBs. The two LSBs can be programmed using the DWLSB register. Display Window LSB register 12 Hex DWLSB Read/Write 32 Hex Bit Symbol 7 28 Description Bit Symbol Register Name Address Mnemonic Type Reset Value Display Row End Register 0F Hex DROWE Read/Write FB Hex Bit Symbol Confidential 7:0 Bit Bit Symbol Register Name Address Mnemonic Type Reset Value Bit Description Register Name Address Mnemonic Type Bit Register Name Address Mnemonic Type Reset Value Description Reserved 6 SwLsb Assert to increment the value of the scan window start and end row addresses by 1. Clear (the default) to use the raw values. 5 DwCel[1] Use to program bit 1 of the display window's end column address. Default is 1. 4 DwCel[0] Use to program bit 0 of the display window's end column address. Default is 1. 3 DwCSL[1] Use to program bit 1 of the display window's start column address. Default is 0. 2 DwCSL [0] Use to program bit 0 of the display window's start column address. Default is 0. 1 DwERLsb Use to program bit 0 of the display window's end row address. Default is 1. 0 DwSRLsb Use to program bit 0 of the display window's start row address. Default is 0. www.national.com Register Name Address Mnemonic Type Reset Value Bit Integration Time High Register 13 Hex ITIMEH Read/Write 00 Hex. Bit Symbol 7:4 3:0 Register Name Address Mnemonic Type Reset Value Bit Description 7:0 Reserved Itime[11:8] Register Name Address Mnemonic Type Reset Value Program to set the integration time of the array. The value programmed in the register is the number of rows ahead of the selected row to be reset. Bit 7:0 Bit Symbol Itime[7:0] Register Name Address Mnemonic Type Reset Value Program to set the integration time of the array. The value programmed in the register is the number of rows ahead of the selected row to be reset. Row Delay High Register 15 Hex RDELAYH Read/Write 00 Hex. Bit Symbol 7:3 2:0 Rdelay[10:8] Register Name Address Mnemonic Type Reset Value Bit 7:0 Use to program the MSBs of the row delay. Row Delay Low Register 16 Hex RDELAYL Read/Write 00 Hex Bit Symbol Rdelay[7:0] Register Name Address Mnemonic Type Reset Value 7:0 Use to program the LSBs of the row delay. Frame Delay High Register 17 FDELAYH Read/Write 00 Hex Bit Symbol 7:4 3:0 Bit Symbol OffsetVol Bit Description Description This register defines the voltage level appearing on the offset_ctrl pin. Offset Compensation Register 1 22 Hex OCR1 Read/Write 00 Hex Bit Symbol OffsetVol Register Name address Mnemonic Type Reset Value Use to program the overall video gain. 00hex corresponds to a gain of 0dB while 3Fhex corresponds to a gain of 15dB. Steps are in linear increments. Offset Compensation Register 0 1FHex OCR0 Read/Write 00 Hex Description 7:0 Bit VidGain Register Name address Mnemonic Type Reset Value Bit Description Reserved Description Reserved Use to program the LSBs of the frame delay. Bit Symbol Register Name address Mnemonic Type Reset Value Bit Description Video Gain Register 19 Hex VGAIN Read/Write 00 Hex Description 7:0 Bit FDelay [7:0] 7:6 5:0 Bit Bit Symbol Register Name Address Mnemonic Type Reset Value Integration Time Low Register 14 Hex ITIMEL Read/Write 00 Hex. Frame Delay Low Register 18 Hex FDELAYL Read/Write 00 Hex Description This register defines the voltage level appearing on the offset_ctrl pin. Offset Compensation Register 2 25 Hex OCR2 Read/Write 00 Hex Bit Symbol OffsetVol Description This register defines the voltage level appearing on the offset_ctrl pin. Reserved FDelay[11:8] Confidential Use to program the MSBs of the frame delay. 29 www.national.com LM9627 Register Set (continued) LM9627 Register Set (continued) Register Name Black Level Register Address 26 Hex Mnemonic BLCOEFF Type Read/Write Reset Value 00 Hex Bit 7:0 Bit Symbol Alpha[7:0] Register Name Address Mnemonic Type Reset Value Bit 7:0 BpT0 [11:4] Bit 7:4 Bit Symbol BpT0 [3.0] Exponential averaging coefficient for black pixels Description Use to program the MSBs of the bad pixel correction threshold 0. Description Use to program the LSBs of the bad pixel correction threshold 0. Reserved Register Name Address Mnemonic Type Reset Value Bit Threshold 1 High Register 29 Hex BPTH1H Read/Write 00 Hex Bit Symbol THR1[11.4] Register Name Address Mnemonic Type Reset Value Bit 7:4 Description Threshold 0 Low Register 28 Hex BPTH0L Read/Write 00 Hex 3:0 7:0 Coefficient Threshold 0 High Register 27 Hex BPTH0H Read/Write 00 Hex. Bit Symbol Register Name Address Mnemonic Type Reset Value Compensation THR1 [3.0] Confidential Use to program the MSBs of the bad pixel correction threshold 1. Threshold 1 Low Register 2A Hex BPTH1L Read/Write 00 Hex Bit Symbol 3:0 Description Description Use to program the LSBs of the bad pixel correction threshold 1. Reserved 30 www.national.com LM9627 Timing Information 1.0 DIGITAL VIDEO PORT MASTER MODE TIMING pclk hsync t2 t1 d[11:0] P0 Pn P1 t3 Figure 49. Row Timing Diagram pclk vsync t6 t5 R2 hsync Rn R3 t1 t2 Figure 50. Frame Timing pclk vsync t6 t5 hsync Fdelayn-2 Fdelayn-1 F delayn R0 R1 R2 Rn t2 t1 Inter Frame Delay Frame (n) Figure 51. Frame Delay Timing (With Inter Frame Delay). Label Descriptions Min Typ Max t0 pclk period 74.4ns 83.3ns 1.0s t1 t2 t3 t5 t6 Note a: Note b: hsync low level mode pulse mode (116-HsyncAdjust) *pclk 16 * pclk (see note a & b) hsync high level mode pulse mode (664 -HsyncAdjust) *pclk 764 * pclk (see note a & b) first valid pixel data after hsync active HsyncAdjust * pclk (see note a & b) vsync low level mode pulse mode 116 *pclk 16 * pclk vsync high level mode pulse mode (FN Hclk - 116) * pclk 16 * pclk (see note a & b) (see note a & b) See Frame Rate Programming section for more details See Digital Video Port Registers for more details Confidential 31 www.national.com LM9627 Timing Information (continued) d[11:0] hsync vsync pclk t1 t2 Figure 52. d[11:0], hsync & vsync to Active High pclk Timing d[11:0] hsync vsync pclk t3 t4 Figure 53. d[11:0], hsync & vsync to Active Low pclk Timing The following specifications apply for all supply pins = +3.3V and C L = 10pF unless otherwise noted. Boldface limits apply for TA = TMIN to T MAX: all other limits TA = 25o C (Note 7) Label Descriptions t1 Rising pclk to Rising hsync, vsync or d[11:0] 25ns t2 Rising pclk to Falling hsync, vsync or d[11:0] 23ns t3 Falling pclk to rising hsync, vsync or d[11:0] 25ns t4 Falling pclk to falling hsync, vsync or d[11:0] 23ns Confidential Min 32 Typ Max www.national.com LM9627 Timing Information (continued) 2.0 DIGITAL VIDEO PORT SLAVE MODE TIMING t1 t3 trigger row n hsync trigger row n+1 t2 d[11:0] P652 P653 P640 P654 P652 P1 P653 P654 P655 mclk Row n-1 Row n Figure 54. Slave Mode Row Trigger and Readout Timing trigger last row in frame n hsync t5 trigger Frame n+1 vsync mclk t4 Figure 55. Slave Mode d[11:0], hsync & vsync to pclk Timing d[11:0] mclk t6 Figure 56. Rising Edge of mclk to Valid Pixel Data The following specifications apply for all supply pins = +3.0V & CL = 10pF unless otherwise noted. Boldface limits apply for TA = TMIN to T MAX: all other limits TA = 25o C (Note 7) Label Descriptions Min t1 Pulse width of row trigger 2 * mclk t2 First pixel out after rising edge of row trigger 124 * mclk t3 Minimum time between row triggers. 780 * mclk t4 Max time to assert next frame trigger after last row trigger. t5 Pulse width of Frame trigger t6 Time to valid pixel data after rising edge of mclk Confidential Typ Max 124 * mclk 96 * mclk 2 * mclk 44ns 33 www.national.com LM9627 Timing Information (continued) 3.0 DIGITAL VIDEO PORT SINGLE FRAME CAPTURE (SNAPSHOT MODE) TIMING t1 snapshot or FTriggerNow irq FTriggerEn extsync or FtSync FtBusy t4 t3 t2 Figure 57. Snapshot Mode Timing With External Shutter t1 snapshot or FTriggerNow irq FTriggerEn extsync or FtSync FtBusy t3 t2 t4 Figure 58. Snapshot Timing Without External Shutter Label Descriptions Equation t1 Minimum Snapshot Trigger Pulse Width 2 * mclk (see notes a & b) t2 Minimum time from Snapshot Pulse to extsync FN Hclk (see notes a & b) t3 Array Integration Time FN Hclk (see notes a & b) t4 Pixel Read Out FN Hclk Note a: Note b: (see notes a & b) See 7.0Frame Rate Programming section for more details See Snapshot Mode for more details Confidential 34 www.national.com LM9627 Timing Information (continued) 4.0 SERIAL BUS TIMING Sr Sr P tfDA tfDA SDA tSU;STA tHD;DAT tHD;STA t SU;STO tSU;DAT SCLK trCL trCL1 trCL trCL1 = Rp resistor pull-up tHIGH tLOW tLOW (1) tHIGH = MCS current source pull-up (1) Rising edge of the first SCLK pulse after an acknowledge bit. Figure 59. I 2 C Compatible Serial Bus Timing. The following specifications apply for all supply pins = +3.3V, CL = 10pF, and sclk = 400KHz unless otherwise noted. Boldface limits apply for TA = TMIN to T MAX: all other limits T A = 25o C (Note 7) Confidential PARAMETER SYMBOL MIN MAX UNIT sclk clock frequency fSCLH 0 400 KHz Set-up time (repeated) START condition tSU;STA 0.6 - S Hold time (repeated) START condition tHD;STA 0.6 - S LOW period of the sclk clock tL O W 1.3 - S HIGH period of the sclk clock tHIGH 0.6 - S Data set-up time tSU;DAT 180 - nS Data hold time tHD;DAT 0 0.9 S Set-up time for STOP condition tSU;STO 0.6 Capacitive load for sda and sclk lines Cb 35 S 400 pF www.national.com LM9627 Array Mechanical Information .440 +/-.005 TYP [11.18 +/- 0.12] .040 +/-.003 TYP [1.02 +/- 0.07] 43 .085 +/-.010 [2.16 +/- 0.25] 48 .060 +.010 TYP -.005 [1.52 + 0.25] [- 0.12] 6 1 7 42 distance from pixel (die surface) to top surface of glass lid= 0.894 mm R.0075 +/-.0050 [0.191+/- 0.127] TYP .020 +/-.003 [0.51 +/- 0.07] TYP 0.328 [8.325] Note 3 31 18 30 .040 +/-.007 TYP [1.02 +/- 0.17] 19 0.281 [7.131] Note 3 Optical Center of Sensor Array (4X R.0075) [0.19] .560 +.012 -.005 [14.22 + 0.30] [ - 0.12] .102 MAX [2.58] Notes: 1. Controlling dimensions are in inches, values in [] are in millimeters 2. All Exposed metallized areas shall be gold plated 60 micro-inches [1.52 micrometers] minimum thickness over nickel plate 3. Reference dimensions only. Tolerance will depend on die placement [+/-0.1 mm]. 4. Reference JEDEC registration MS-009, variation AF issue A, dated 9/29/1980. Confidential 36 www.national.com LM9627 Color CMOS Image Sensor VGA 30 FPS LIFE SUPPORT POLICY NATIONAL'S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT AND GENERAL COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein: 1. Life support devices or systems are devices or systems which, (a) are intended for surgical implant into the body, or (b) support or sustain life, and whose failure to perform when properly used in accordance with instructions for use provided in the labeling, can be reasonably expected to result in a significant injury to the user. National Semiconductor Corporation Americas Tel: 1-800-272-9959 Fax: 1-800-737-7018 Email: support @ nsc.com www.national.com National Semiconductor Europe Fax: +49 (0) 1 80-530 85 86 Email: europe.support @ nsc.com Deutsch Tel: +49 (0) 69 9508 6208 English Tel: +44 (0) 870 24 0 2171 Francais Tel: +33 (0) 1 41 91 8790 2. A critical component is any component of a life support device or system whose failure to perform can be reasonably expected to cause the failure of the life support device or system, or to affect its safety or effectiveness. National Semiconductor Asia Pacific Customer Response Group Tel: 65-2544466 Fax: 65-2504466 Email: ap.support@nsc.com National Semiconductor Japan Ltd. Tel: 81-3-5639-7560 Fax: 81-3-5639-7507 National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.