Dual-Channel, 14-Bit, CCD Signal Processor with Precision Timing Core AD9974 FEATURES GENERAL DESCRIPTION 1.8 V analog and digital core supply voltage Correlated double sampler (CDS) with −3 dB, 0 dB, +3 dB, and +6 dB gain 6 dB to 42 dB, 10-bit variable gain amplifier (VGA) 14-bit, 65 MHz analog-to-digital converter (ADC) Black level clamp with variable level control Complete on-chip timing generator Precision Timing core with 240 ps resolution @ 65 MHz On-chip 3 V horizontal and RG drivers 100-lead, 9 mm × 9 mm, 0.8 mm pitch, CSP_BGA package Internal low dropout (LDO) regulator circuitry The AD9974 is a highly integrated, dual-channel, chargecoupled device (CCD) signal processor for high speed digital video camera applications. Each channel is specified at pixel rates of up to 65 MHz. The AD9974 consists of a complete analog front end (AFE) with analog-to-digital conversion, combined with a programmable timing driver. The Precision Timing™ core allows adjustment of high speed clocks with approximately 240 ps resolution at 65 MHz operation. Each AFE includes black level clamping, CDS, VGA, and a 65 MSPS, 14-bit ADC. The timing driver provides the high speed CCD clock drivers for the RG_A, RG_B, H1_A to H4_A, and H1_B to H4_B outputs. A 3-wire serial interface is used to program each channel of the AD9974. APPLICATIONS Professional HDTV camcorders Professional/high end digital cameras Broadcast cameras Industrial high speed cameras Available in a space-saving, 9 mm × 9 mm, CSP_BGA package, the AD9974 is specified over an operating temperature range of −25°C to +85°C. FUNCTIONAL BLOCK DIAGRAM REFT_A REFB_A REFT_B REFB_B VREF_A VREF_B AD9974 CCDINP_A 14 CDS CCDINM_A DOUT_A ADC VGA 6dB TO 42dB –3, 0, +3, +6dB CLAMP CLAMP 6dB TO 42dB –3, 0, +3, +6dB CCDINP_B CDS 14 ADC VGA DOUT_B CCDINM_B 1.8V OUTPUT LDO A 1.8V OUTPUT LDO B RG_A RG_B H1_B TO H4_B PRECISION TIMING CORE 4 4 CLI_A CLI_B HORIZONTAL DRIVERS SYNC GENERATOR INTERNAL REGISTERS HD_A VD_A HD_B VD_B SL_A SDATA_A SL_B SDATA_B SCK_A SCK_B 05955-001 H1_A TO H4_A INTERNAL CLOCKS Figure 1. Rev. A Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other rights of third parties that may result from its use. Specifications subject to change without notice. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. Trademarks and registered trademarks are the property of their respective owners. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781.329.4700 www.analog.com Fax: 781.461.3113 ©2009 Analog Devices, Inc. All rights reserved. AD9974 TABLE OF CONTENTS Features .............................................................................................. 1 Precision Timing High Speed Timing Core ............................. 15 Applications ....................................................................................... 1 Horizontal Clamping and Blanking ......................................... 18 General Description ......................................................................... 1 Complete Field—Combining H-Patterns ............................... 25 Functional Block Diagram .............................................................. 1 Mode Registers ........................................................................... 26 Revision History ............................................................................... 2 Horizontal Timing Sequence Example.................................... 28 Specifications..................................................................................... 3 Analog Front End Description and Operation ...................... 29 Channel-to-Channel Specifications ........................................... 3 Applications Information .............................................................. 33 Timing Specifications .................................................................. 4 Recommended Power-Up Sequence ....................................... 33 Digital Specifications ................................................................... 5 Standby Mode Operation .......................................................... 36 Analog Specifications ................................................................... 6 CLI Frequency Change .............................................................. 36 Absolute Maximum Ratings............................................................ 8 Circuit Configuration ................................................................ 37 Thermal Characteristics .............................................................. 8 Grounding and Decoupling Recommendations .................... 37 ESD Caution .................................................................................. 8 3-Wire Serial Interface Timing ..................................................... 39 Pin Configuration and Function Descriptions ............................. 9 Layout of Internal Registers ...................................................... 40 Typical Performance Characteristics ........................................... 11 Updating of Register Values ...................................................... 41 Equivalent Input/Output Circuits ................................................ 12 Complete Register Listing ............................................................. 42 Terminology .................................................................................... 13 Outline Dimensions ....................................................................... 50 Theory of Operation ...................................................................... 14 Ordering Guide .......................................................................... 50 Programmable Timing Generation .............................................. 15 REVISION HISTORY 10/09—Revision A: Initial Version Changes to Table 1 ............................................................................ 3 Changes to Table 3 ............................................................................ 4 Changes to Pin Function Descriptions Table ............................... 9 Changes to Figure 11 ...................................................................... 12 Changes to Individual HBLK Pattern Section ............................ 20 Changes to Table 14 ........................................................................ 25 Added Example Register Setting for Power-Up Section ........... 34 Added Additional Restrictions Section ....................................... 35 Changes to Table 2 .......................................................................... 36 Changes to 3 V System Compatibility Section ........................... 37 Changes to Grounding and Decoupling Recommendations Section ............................................................ 37 Changes to Table 30 ........................................................................ 48 Changes to Table 31 ........................................................................ 49 Changes to Ordering Guide .......................................................... 50 Rev. A | Page 2 of 52 AD9974 SPECIFICATIONS X = A = B, unless otherwise noted. Table 1. Parameter TEMPERATURE RANGE Operating Storage POWER SUPPLY VOLTAGE AVDD_X (AFE, Timing Core) RGVDD_X (RG_X Driver) HVDD_X (H1_X to H4_X Drivers) DVDD_X (All Other Digital) DRVDD_X (Parallel Data Output Drivers) IOVDD_X (I/O Supply Without the Use of LDO) POWER SUPPLY CURRENTS—65 MHz OPERATION AVDD_X (1.8 V) RGVDD_X (3.3 V, 20 pF RG Load) HVDD_X 1 (3.3 V, 200 pF Total Load on H1 to H4) DVDD_X (1.8 V) DRVDD_X (3.0 V) IOVDD_X (1.8 V) POWER SUPPLY CURRENTS—STANDBY MODE OPERATION Reference Standby Total Shutdown LDO 2 IOVDD_X (I/O Supply When Using LDO) Output Voltage Output Current CLOCK RATE (CLI) Min Typ −25 −65 1.6 2.7 2.7 1.6 1.6 1.6 8 1.8 3.3 3.3 1.8 3.0 1.8 Max Unit +85 +150 °C °C 2.0 3.6 3.6 2.0 3.6 3.6 V V V V V V 55 5 40 15 3 2 mA mA mA mA mA mA 10 0.5 mA mA 3.0 1.85 60 V V mA MHz 100 65 1 The total power dissipated by the HVDD (or RGVDD) supply can be approximated as follows: Total HVDD Power = [CLOAD × HVDD × Pixel Frequency] × HVDD. Reducing the capacitive load and/or reducing the HVDD supply reduces the power dissipation. CLOAD is the total capacitance seen by all H-outputs. 2 LDO should be used to supply only AVDD and DVDD. CHANNEL-TO-CHANNEL SPECIFICATIONS X = A = B, TMIN to TMAX, AVDD_X = DVDD_X = 1.8 V, fCLI = 65 MHz, typical timing specifications, unless otherwise noted. Table 2. Parameter LINEARITY MISMATCH 1 CROSSTALK ERROR Channel A to Channel B Channel B to Channel A 1 Min Typ <0.5 −82 −82 Max Unit % dB dB Test Conditions/Comments Absolute value above 1⁄16 of maximum output code CDS = 0 dB Full-scale step applied to Channel A while measuring response on Channel B Full-scale step applied to Channel B while measuring response on Channel A See the Terminology section for further measurement explanation. Rev. A | Page 3 of 52 AD9974 TIMING SPECIFICATIONS X = A = B, CL = 20 pF, AVDD_X = DVDD_X = 1.8 V, fCLI = 65 MHz, unless otherwise noted. Table 3. Parameter MASTER CLOCK (CLI) CLI Clock Period (tCONV) CLI High/Low Pulse Width (tADC) Delay from CLI Rising Edge to Internal Pixel Position 0 (tCLIDLY) AFE SHP Rising Edge to SHD Rising Edge (tS1) AFE Pipeline Delay CLPOB Pulse Width (Programmable) (tCOB) 1 HD Pulse Width VD Pulse Width SERIAL INTERFACE Maximum SCK Frequency (fSCLK) SL to SCK Setup Time (tLS) SCK to SL Hold Time (tLH) SDATA Valid to SCK Rising Edge Setup (tDS) SCK Rising Edge to SDATA Valid Hold (tDH) H-COUNTER RESET SPECIFICATIONS HD Pulse Width VD Pulse Width VD Falling Edge to HD Falling Edge(tVDHD) HD Falling Edge to CLI Rising Edge(tHDCLI) CLI Rising Edge to SHPLOC (Internal Sample Edge) (tCLISHP) TIMING CORE SETTING RESTRICTIONS Inhibited Region for SHP Edge Location (tSHPINH) (See Figure 21) 2 Inhibited Region for SHP or SHD with Respect to H-Clocks (See Figure 21) 3, 4, 5, 6 RETIME = 0, MASK = 0 (tSHDINH) RETIME = 0, MASK = 1 (tSHDINH) RETIME = 1, MASK = 0 (tSHPINH) RETIME = 1, MASK = 1 (tSHPINH) Inhibited Region for DOUTPHASE Edge Location (tDOUTINH) (See Figure 21) Min 15.38 6.9 6.9 2 tCONV 1 HD period Typ Max Unit 7.7 5 8.9 ns ns ns 7.7 16 20 8.5 ns Cycles Pixels ns ns Comments See Figure 17 See Figure 21 See Figure 22 See Figure 52 40 10 10 10 10 MHz ns ns ns ns tCONV 1 HD period 0 3 3 VD period − tCONV tCONV − 2 tCONV − 2 ns ns ns ns ns 50 64/0 Edge location H × NEGLOC − 15 H × POSLOC − 15 H × NEGLOC − 15 H × POSLOC − 15 SHDLOC + 0 H × NEGLOC − 0 H × POSLOC − 0 H × NEGLOC − 0 H × POSLOC − 0 SHDLOC + 15 Edge location Edge location Edge location Edge location Edge location See Figure 49 1 Minimum CLPOB pulse width is for functional operation only. Wider typical pulses are recommended to achieve good clamp performance. Only applies to slave mode operation. The inhibited area for SHP is needed to meet the timing requirements for tCLISHP for proper H-counter reset operation. 3 When 0x34[2:0] HxBLKRETIME bits are enabled, the inhibit region for SHD location changes to inhibit region for SHP location. 4 When sequence register 0x09[23:21] HBLK masking registers are set to 0, the H-edge reference becomes H × NEGLOC. 5 The H-clock signals that have SHP/SHD inhibit regions depend on the HCLK mode: Mode 1 = H1, Mode 2 = H1, H2, and Mode 3 = H1, H3. 6 These specifications apply when H1POL, H2POL, RGPOL, and HLPOL are all set to 1 (default setting). 2 Rev. A | Page 4 of 52 AD9974 DIGITAL SPECIFICATIONS X = A = B, IOVDD_X = 1.6 V to 3.6 V, RGVDD_X = HVDD_X = 2.7 V to 3.6 V, CL = 20 pF, TMIN to TMAX, unless otherwise noted. Table 4. Parameter LOGIC INPUTS High Level Input Voltage (VIH) Low Level Input Voltage (VIL) High Level Input Current (IIH) Low Level Input Current (IIL) Input Capacitance (CIN) LOGIC OUTPUTS High Level Output Voltage (VOH) Low Level Output Voltage (VOL) CLI INPUT (CLI_BIAS = 0) High Level Input Voltage (VIHCLI) Low Level Input Voltage (VILCLI) H-DRIVER OUTPUTS High Level Output Voltage at Maximum Current (VOH) Low Level Output Voltage at Maximum Current Maximum Output Current (Programmable) (VOL) Maximum Load Capacitance Min Typ Max IOVDD − 0.6 0.6 10 10 10 IOVDD − 0.5 V V IOVDD/2 − 0.5 V V HVDD − 0.5 0.5 30 Rev. A | Page 5 of 52 Test Conditions/Comments V V μA μA pF 0.5 IOVDD/2 + 0.5 100 Unit V V mA pF IOH = 2 mA IOL = 2 mA AD9974 ANALOG SPECIFICATIONS X = A = B, AVDD_X = 1.8 V, fCLI = 65 MHz, typical timing specifications, TMIN to TMAX, unless otherwise noted. Table 5. Parameter CDS 1 Allowable CCD Reset Transient CDS Gain Accuracy −3 dB CDS Gain 0 dB CDS Gain (Default) 3 dB CDS Gain 6 dB CDS Gain Maximum Input Voltage −3 dB CDS Gain 0 dB CDS Gain (Default) 3 dB CDS Gain 6 dB CDS Gain Allowable OB Pixel Amplitude 0 dB CDS Gain (Default) 6 dB CDS Gain VARIABLE GAIN AMPLIFIER (VGA_X) Gain Control Resolution Gain Monotonicity Low Gain Setting (VGA Code 15, Default) Maximum Gain Setting (VGA Code 1023) BLACK LEVEL CLAMP Clamp Level Resolution Minimum Clamp Level (Code 0) Maximum Clamp Level (Code 1023) ADC (CHN_A and CHN_B) Resolution Differential Nonlinearity (DNL) No Missing Codes Integral Nonlinearity (INL) Full-Scale Input Voltage VOLTAGE REFERENCE Reference Top Voltage (REFT_X) Reference Bottom Voltage (REFB_X) SYSTEM PERFORMANCE VGA Gain Accuracy Low Gain (Code 15) Maximum Gain (Code 1023) Peak Nonlinearity, 500 mV Input Signal Total Output Noise Power Supply Rejection (PSR) 1 Min −3.3 −0.7 2.3 4.9 Typ Max Unit 0.5 0.8 V −2.8 −0.2 2.8 5.4 −2.3 +0.3 3.3 5.9 Test Conditions/Comments VGA gain = 5.6 dB (Code 15, default value) 1.4 1.0 0.7 0.5 V p-p V p-p V p-p V p-p −100 −50 14 −1.0 +200 +100 1024 Guaranteed 6 42 dB dB 1024 0 1023 Steps LSB LSB ±0.5 Guaranteed 5 2.0 Steps +1.2 15 1.4 0.4 5.1 41.3 mV mV 5.6 41.8 0.1 2 48 Measured at ADC output Measured at ADC output Bits LSB LSB V V V 6.1 42.3 0.4 Input signal characteristics are defined as shown in Figure 2. Rev. A | Page 6 of 52 dB dB % LSB rms dB Specifications include entire signal chain 0 dB CDS gain (default) Gain = (0.0359 × code) + 5.1 dB 12 dB total gain applied AC-grounded input, 6 dB gain applied Measured with step change on supply AD9974 MAXIMUM INPUT LIMIT = LESSER OF 2.2V OR (AVDD + 0.3V) +1.8V TYP (AVDD) 800mV MAXIMUM 200mV MAX OPTICAL BLACK PIXEL +1.3V TYP (AVDD – 0.5V) DC RESTORE VOLTAGE 1V MAXIMUM INPUT SIGNAL RANGE (0dB CDS GAIN) 0V (AVSS) MINIMUM INPUT LIMIT (AVSS – 0.3V) Figure 2. Input Signal Characteristics Rev. A | Page 7 of 52 05957-002 500mV TYP RESET TRANSIENT AD9974 ABSOLUTE MAXIMUM RATINGS Ratings apply to both Channel A and Channel B, unless otherwise noted. Table 6. Parameter AVDD to AVSS DVDD to DVSS DRVDD to DRVSS IOVDD to DVSS HVDD to HVSS RGVDD to RGVSS Any VSS RG Output to RGVSS H1 to H4, HL Output to HVSS SCK, SL, SDI to DVSS REFT, REFB, CCDINM, CCDINP to AVSS Junction Temperature Lead Temperature (10 sec) Rating −0.3 V to +2.2 V −0.3 V to +2.2 V −0.3 V to +3.9 V −0.3 V to +3.9 V −0.3 V to +3.9 V −0.3 V to +3.9 V −0.3 V to +0.3 V −0.3 V to RGVDD + 0.3 V −0.3 V to HVDD + 0.3 V −0.3 V to IOVDD + 0.3 V −0.2 V to AVDD + 0.2 V 150°C 350°C Stresses above those listed under Absolute Maximum Ratings may cause permanent damage to the device. This is a stress rating only; functional operation of the device at these or any other conditions above those indicated in the operational section of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability. THERMAL CHARACTERISTICS θJA is measured using a 4-layer PCB with the exposed paddle soldered to the board. Table 7. Thermal Resistance Package Type 100-Lead, 9 mm × 9 mm, CSP_BGA ESD CAUTION Rev. A | Page 8 of 52 θJA 38.3 Unit °C/W AD9974 PIN CONFIGURATION AND FUNCTION DESCRIPTIONS AD9974 TOP VIEW (Not to Scale) A1 CORNER INDEX AREA 1 2 3 4 5 6 7 8 9 10 A B C D E F G H 05955-003 J K Figure 3. Pin Configuration Table 8. Pin Function Descriptions Ball Location B2 C2 D2 C1 D1 A1 F4 F3 D4 D3 B4 J2 K3 E3 E4 C3 C4 B3 Mnemonic SL_A SDATA_A SCK_A REFT_A REFB_A CCDINM_A H1_A H2_A H3_A H4_A RG_A DRVSS_A DRVDD_A HVSS_A HVDD_A RGVSS_A RGVDD_A IOVDD_A Type 1 DI DI DI AO AO AI DO DO DO DO DO P P P P P P P A4 B1 A2 F2 F1 E2 E1 B8 C8 A5 A6 D8 C7 D7 A7 F10 F9 CLI_A AVSS_A CCDINP_A DVSS_A DVDD_A VD_A HD_A SL_B SDATA_B LDO_OUT_A CCDINM_B SCK_B REFT_B REFB_B CCDINP_B H1_B H2_B DI P AI P P DI DI DI DI P AI DI AO AO AI DO DO Description 3-Wire Serial Load for Channel A. 3-Wire Serial Data for Channel A. 3-Wire Serial Clock for Channel A. Reference with Top Decoupling for Channel A. Decouple with 0.1 μF to AVSS_A. Reference with Bottom Decoupling for Channel A. Decouple with 0.1 μF to AVSS_A. Analog Input for Channel A Image Sensor Signal. CCD Horizontal Clock 1 for Channel A. CCD Horizontal Clock 2 for Channel A. CCD Horizontal Clock 3 for Channel A. CCD Horizontal Clock 4 for Channel A. CCD Reset Gate Clock for Channel A. Digital Driver Ground for Channel A. Digital Driver Supply for Channel A: 1.8 V or 3.0 V. H1_A to H4_A Driver Ground for Channel A. H1_A to H4_A Driver Supply for Channel A: 3.0 V. RG_A Driver Ground for Channel A. RG_A Driver Supply for Channel A: 3.0 V. Digital I/O Supply: 1.8 V or 3.0 V (HD, VD, SL, SCK, SDATA) and LDO Input (3.0 V Only) When LDO Is Used. Master Clock Input for Channel A. Analog Ground for Channel A. Analog Input for Channel A Image Sensor Signal. Digital Ground for Channel A. Digital Supply for Channel A: 1.8 V. Vertical Sync Pulse for Channel A. Horizontal Sync Pulse for Channel A. 3-Wire Serial Load for Channel B. 3-Wire Serial Data for Channel B. 1.8 V LDO Output from Channel A. Analog Input for Channel B Image Sensor Signal. 3-Wire Serial Clock for Channel B. Reference with Top Decoupling for Channel B. Decouple with 0.1 μF to AVSS_B. Reference with Bottom Decoupling for Channel B. Decouple with 0.1 μF to AVSS_B. Analog Input for Channel B Image Sensor Signal. CCD Horizontal Clock 1 for Channel B. CCD Horizontal Clock 2 for Channel B. Rev. A | Page 9 of 52 AD9974 Ball Location D10 D9 B10 J8 K9 E9 E10 C9 C10 B9 Mnemonic H3_B H4_B RG_B DRVSS_B DRVDD_B HVSS_B HVDD_B RGVSS_B RGVDD_B IOVDD_B Type 1 DO DO DO P P P P P P P A10 B7 A8 F8 F7 E8 E7 A3 G1 H1 J1 K1 G2 H2 K2 G3 H3 J3 K4 J4 H4 G4 B5, C5, D5, E5, F5, G5, H5, J5, K5, B6, C6, D6, E6, F6, G6, H6, J6, K6 A9 G7 H7 J7 K7 G8 H8 K8 G9 H9 J9 K10 J10 H10 G10 LDO_OUT_B AVSS_B AVDD_B DVSS_B DVDD_B VD_B HD_B AVDD_A D0_A D1_A D2_A D3_A D4_A D5_A D6_A D7_A D8_A D9_A D10_A D11_A D12_A D13_A GND P P P P P DI DI P DO DO DO DO DO DO DO DO DO DO DO DO DO DO P Description CCD Horizontal Clock 3 for Channel B. CCD Horizontal Clock 4 for Channel B. CCD Reset Gate Clock for Channel B. Digital Driver Ground for Channel B. Digital Driver Supply for Channel B: 1.8 V or 3.0 V. H1_B to H4_B Driver Ground for Channel B. H1_B to H4_B Driver Supply for Channel B: 3.0 V. RG_B Driver Ground for Channel B. RG_B Driver Supply for Channel B: 3.0 V. Digital I/O Supply: 1.8 V or 3.0 V (HD, VD, SL, SCK, SDATA) and LDO Input (3.0 V Only) When LDO Is Used. 1.8 V LDO Output from Channel B. Analog Ground for Channel B. Analog Supply for Channel B: 1.8 V. Digital Ground for Channel B. Digital Supply for Channel B: 1.8 V. Vertical Sync Pulse for Channel B. Horizontal Sync Pulse for Channel B. Analog Supply for Channel A: 1.8 V. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Data Outputs Channel A. Ground Connection. CLI_B D0_B D1_B D2_B D3_B D4_B D5_B D6_B D7_B D8_B D9_B D10_B D11_B D12_B D13_B DI DO DO DO DO DO DO DO DO DO DO DO DO DO DO Master Clock Input for Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. Data Outputs Channel B. 1 AI = analog input, AO = analog output, DI = digital input, DO = digital output, P = power. Rev. A | Page 10 of 52 AD9974 TYPICAL PERFORMANCE CHARACTERISTICS 10 250 8 200 6 TOTAL POWER LSB POWER (mW) 4 150 3.3V SUPPLIES 100 2 0 –2 –4 50 1.8V SUPPLIES 30 35 40 45 50 55 60 65 SAMPLE RATE (MHz) –8 0 2k 8k 10k 12k 14k 16k Figure 7. Integral Nonlinearity 180 1.0 160 0.8 140 0.6 120 INL MISMATCH (%) 100 80 60 40 0.4 0.2 0 –0.2 –0.4 –0.6 20 0 5 10 15 20 25 30 35 40 45 VGA GAIN (dB) 05955-006 –0.8 0 Figure 5. RMS Output Noise vs. VGA Gain 0.8 0.6 0.4 0.2 0 –0.2 –0.4 –0.6 –1.0 4k 6k 8k 10k 12k ADC OUTPUT CODE 14k 16k 05955-007 –0.8 2k 3k 5k 7k 9k 11k 13k 15k ADC OUTPUT CODE Figure 8. Linearity Mismatch vs. ADC Output Code 1.0 0 –1.0 1k Figure 6. Differential Nonlinearity Rev. A | Page 11 of 52 05955-009 RMS OUTPUT NOISE (LSB) 6k ADC OUTPUT CODE Figure 4. Power vs. Sample Rate LSB 4k 05955-008 25 05955-005 0 20 –6 AD9974 EQUIVALENT INPUT/OUTPUT CIRCUITS IOVDD AVDD 330Ω R CLI 100kΩ AVSS 05955-011 AVSS 05955-010 + AVSS Figure 11. CLI Input, Register 0x15[0] = 1 Enables the Bias Circuit Figure 9. CCDIN Input HVDD OR RGVDD DATA IOVDD OUTPUT ENABLE HVSS OR RGVSS Figure 10. Digital Inputs Figure 12. H1 to H4 and RG Outputs Rev. A | Page 12 of 52 05955-013 DVSS 05955-012 330Ω AD9974 TERMINOLOGY MAX Differential Nonlinearity (DNL) An ideal ADC exhibits code transitions that are exactly 1 LSB apart. DNL is the deviation from this ideal value. Therefore, every code must have a finite width. No missing codes guaranteed to 14-bit resolution indicates that all 16,384 codes, each for its respective input, must be present over all operating conditions. Total Output Noise The rms output noise is measured using histogram techniques. The standard deviation of the ADC output codes is calculated in LSB and represents the rms noise level of the total signal chain at the specified gain setting. The output noise can be converted to an equivalent voltage using the relationship 1 LSB = (ADC Full Scale/2n Codes) OUTPUT CODE (LSB) CHANNEL A INL A(X) INL B(X) OUTPUT (X) MAX/16 0 0 X FS INPUT VOLTAGE Figure 13. Linearity Mismatch Definition Power Supply Rejection (PSR) The PSR is measured with a step change applied to the supply pins. The PSR specification is calculated from the change in the data outputs for a given step change in the supply voltage. Crosstalk The crosstalk is measured while applying a full-scale step to one channel and measuring the interference on the opposite channel. ⎛ Interference (LSB) ⎞ ⎟⎟ Crosstalk (dB) = 20 × log ⎜⎜ 16,384 ⎠ ⎝ where n is the bit resolution of the ADC. For the AD9974, 1 LSB is approximately 122.0 μV. Linearity Mismatch The linearity mismatch is calculated by taking the difference in INL of the two channels at Input X, and then expressing the difference as a percentage of the output code at X. The values given in Table 2 are obtained over the range of 1⁄16 and maximum of the output code. The general trend is for the linearity mismatch to decrease as the output approaches the maximum code, as shown in Figure 8. Linearity Mismatch (%) = IDEAL 05955-004 Peak Nonlinearity Peak nonlinearity, a full signal chain specification, refers to the peak deviation of the output of the AD9974 from a true straight line. The point used as zero scale occurs 0.5 LSB before the first code transition. Positive full scale is defined as a level 1 LSB and 0.5 LSB beyond the last code transition. The deviation is measured from the middle of each particular output code to the true straight line. The error is then expressed as a percentage of the 2 V ADC full-scale signal. The input signal is always appropriately gained up to fill the ADC full-scale range. CHANNEL B INL A( X ) − INL B( X ) Output Code( X ) Rev. A | Page 13 of 52 AD9974 THEORY OF OPERATION All AD9974 clocks are synchronized with VD and HD inputs. All of the AD9974 horizontal pulses (CLPOB, PBLK, and HBLK) are programmed and generated internally. V DRIVER V1 > Vx, VSG1 > VSGx, SUBCK H1_A TO H4_A, RG_A H1_B TO H4_B, RG_B The H-drivers for H1 to H4 and RG are included in the AD9974, allowing these clocks to be directly connected to the CCD. An H-driver voltage of 3 V is supported in the AD9974. DOUT_A DOUT_B CCDINP_A CCDINM_A CCD CCDINP_B AD9974 INTEGRATED HD_A, VD_A, AFE + TD HD_B, VD_B DIGITAL IMAGE PROCESSING ASIC CCDINM_B 05955-014 CLI_A, CLI_B SERIAL INTERFACE Figure 15 and Figure 16 show the maximum horizontal and vertical counter dimensions for the AD9974. All internal horizontal and vertical clocking is controlled by these counters, which specify line and pixel locations. Maximum HD length is 8191 pixels per line, and maximum VD length is 8191 lines per field. MAXIMUM COUNTER DIMENSIONS Figure 14. Typical Application 13-BIT HORIZONTAL = 8192 PIXELS MAX 13-BIT VERTICAL = 8192 LINES MAX 05955-015 Figure 14 shows the typical system block diagram for the AD9974. The charge-coupled device (CCD) output is processed by the analog front-end (AFE) circuitry of the AD9974, consisting of a CDS, VGA, black level clamp, and ADC. The digitized pixel information is sent to the digital image processor chip, which performs the postprocessing and compression. To operate the CCD, all CCD timing parameters are programmed into the AD9974 from the system ASIC through the 3-wire serial interface. From the system master clock, CLI_X, which is provided by the image processor or external crystal, the AD9974 generates the horizontal clocks of the CCD and all internal AFE clocks. Figure 15. Vertical and Horizontal Counters MAX VD LENGTH IS 8192 LINES VD MAX HD LENGTH IS 8192 PIXELS HD 05955-016 CLI Figure 16. Maximum VD/HD Dimensions Rev. A | Page 14 of 52 AD9974 PROGRAMMABLE TIMING GENERATION Using a 65 MHz CLI frequency, the edge resolution of the Precision Timing core is approximately 240 ps. If a 1× system clock is not available, it is possible to use a 2× reference clock by programming the CLIDIVIDE register (Address 0x0D). The AD9974 then internally divides the CLI frequency by 2. PRECISION TIMING HIGH SPEED TIMING CORE The AD9974 generates flexible high speed timing signals using the Precision Timing core. This core, composed of the Reset Gate RG, Horizontal Driver H1 to Horizontal Driver H4, and SHP/SHD sample clocks, is the foundation for generating the timing for both the CCD and the AFE. A unique architecture makes it routine for the system designer to optimize image quality by providing precise control over the horizontal CCD readout and the AFE correlated double sampling. High Speed Clock Programmability Figure 18 shows when the high speed clocks, RG, H1 to H4, SHP, and SHD, are generated. The RG pulse has programmable rising and falling edges and can be inverted using the polarity control. The H1 and H2 horizontal clocks have separate programmable rising and falling edges, as well as separate polarity control. The AD9974 provides additional HCLK-mode programmability, as described in Table 9. Timing Resolution The Precision Timing core uses a master clock input (CLI_X) as a reference. This clock input should be the same as the CCD pixel clock frequency. Figure 17 illustrates how the internal timing core divides the master clock period into 64 steps or edge positions; therefore, the edge resolution of the Precision Timing core is (tCLI/64). For more information on using the CLI input, refer to the Applications Information section. POSITION P[0] P[16] The edge location registers are each six bits wide, allowing the selection of all 64 edge locations. Figure 21 shows the default timing locations for all of the high speed clock signals. P[32] P[48] P[64] = P[0] CLI tCLIDLY 1 PIXEL PERIOD 05955-017 NOTES 1. THE PIXEL CLOCK PERIOD IS DIVIDED INTO 64 POSITIONS, PROVIDING FINE EDGE RESOLUTION FOR HIGH SPEED CLOCKS. 2. THERE IS A FIXED DELAY FROM THE CLI INPUT TO THE INTERNAL PIXEL PERIOD POSITION (tCLIDLY). Figure 17. High Speed Clock Resolution from CLI Master Clock Input 1 2 CCD SIGNAL 3 4 RG 5 6 H1, H3 H2, H4 05955-018 PROGRAMMABLE CLOCK POSITIONS: 1SHP SAMPLE LOCATION. 2SHD SAMPLE LOCATION. 3RG RISING EDGE. 4RG FALLING EDGE. 5H1 RISING EDGE. 6H1 FALLING EDGE. Figure 18. High Speed Clock Programmable Locations (HCLK Mode 1) Rev. A | Page 15 of 52 AD9974 1 2 H1, H3 4 3 H2, H4 05955-019 H1 TO H4 PROGRAMMABLE LOCATIONS: 1H1 RISING EDGE. 2H1 FALLING EDGE. 3H2 RISING EDGE. 4H2 FALLING EDGE. Figure 19. HCLK Mode 2 Operation 1 2 H1 H2 3 4 H3 H4 05955-020 H1 TO H4 PROGRAMMABLE LOCATIONS: 1H1 RISING EDGE. 2H1 FALLING EDGE. 3H3 RISING EDGE. 4H3 FALLING EDGE. Figure 20. HCLK Mode 3 Operation POSITION P[0] P[16] RGr[0] RGf[16] P[32] P[48] P[64] = P[0] CLI RG H1r[0] H1f[32] H1 H2 tS1 CCD SIGNAL SHPLOC[32] SHP tSHPINH SHDLOC[0] tSHDINH SHD DATAPHASEP[32] tDOUTINH NOTES 1. ALL SIGNAL EDGES ARE FULLY PROGRAMMABLE TO ANY OF THE 64 POSITIONS WITHIN ONE PIXEL PERIOD. TYPICAL POSITIONS FOR EACH SIGNAL ARE SHOWN. HCLK MODE 1 IS SHOWN. 2. CERTAIN POSITIONS SHOULD BE AVOIDED FOR EACH SIGNAL, SHOWN ABOVE AS INHIBIT REGIONS. 3. IF A SETTING IN THE INHIBIT REGION IS USED, AN UNSTABLE PIXEL SHIFT CAN OCCUR IN THE HBLK LOCATION OR AFE PIPELINE. Figure 21. High Speed Timing Default Locations Rev. A | Page 16 of 52 05955-021 DOUTPHASEP AD9974 Table 9. HCLK Modes, Selected by HCLKMODE Register (Address 0x23[7:5]) HCLK Mode Mode 1 Mode 2 Register Value 001 010 Mode 3 100 Invalid Selection 000, 011, 101, 110, 111 Description H1 edges are programmable, with H3 = H1 and H2 = H4 = inverse of H1. H1 edges are programmable, with H3 = H1. H2 edges are programmable, with H4 = H2. H1 edges are programmable, with H2 = inverse of H1. H3 edges are programmable, with H4 = inverse of H3. Invalid register settings. Table 10. H1, H2, RGCONTROL, DRVCONTROL, and SAMPCONTROL Register Parameters Parameter Polarity Length (Bits) 1 Range High/low Positive Edge Negative Edge Sample Location Drive Control 6 6 6 3 0 to 63 edge location 0 to 63 edge location 0 to 63 sample location 0 to 7 current steps Description Polarity control for H1/H3 and RG. 0 = no inversion. 1 = inversion. Positive edge location for H1/H3 and RG. Negative edge location for H1/H3 and RG. Sampling location for SHP and SHD. Drive current for H1 to H4 and RG outputs, 0 to 7 steps of 4.3 mA each. CLI tCLIDLY N N+1 N+2 N+3 N+4 N+5 N+6 N+7 N–14 N–13 N–12 N–11 N–10 N+8 N+9 N+10 N+11 N+12 N+13 N+14 N+15 N+16 N+17 N–9 N–8 N–7 N–6 N–5 N–4 N–3 N–2 N–1 N CCDIN SAMPLE PIXEL N SHD (INTERNAL) ADC OUT (INTERNAL) N–17 N–16 N–15 N+1 tDOUTINH DOUTPHASE CLK PIPELINE LATENCY = 16 CYCLES N–16 N–15 N–14 N–13 N–12 N–11 N–10 N–9 N–8 N–7 N–6 N–5 N–4 N–3 N–2 N–1 NOTES 1. EXAMPLE SHOWN FOR SHDLOC = 0. 2. HIGHER VALUES OF SHD AND/OR DOUTPHASE WILL SHIFT DOUT TRANSITION TO THE RIGHT, WITH RESPECT TO CLI LOCATION. N N+1 05955-022 DOUT N–17 Figure 22. Pipeline Delay of AFE Data Outputs H-Driver and RG Outputs Digital Data Outputs In addition to the programmable timing positions, the AD9974 features on-chip output drivers for the RG and H1 to H4 outputs. These drivers are powerful enough to drive the CCD inputs directly. The H-driver and RG-driver current can be adjusted for optimum rise/fall time into a particular load by using the drive strength control registers (Address 0x35). Use the register to adjust the drive strength in 4.3 mA increments. The minimum setting of 0 is equal to off or three-state, and the maximum setting of 7 is equal to 30.1 mA. For maximum system flexibility, the AD9974 uses the DOUTPHASE registers (Address 0x37[11:0]) to select the location for the start of each new pixel data value. Any edge location from 0 to 63 can be programmed. These registers determine the start location of the data output and the DCLK rising edge with respect to the master clock input, CLI_X. The pipeline delay through the AD9974 is shown in Figure 22. After the CCD input is sampled by SHD, there is a 16-cycle delay until the data is available. Rev. A | Page 17 of 52 AD9974 HORIZONTAL CLAMPING AND BLANKING CLPOB and PBLK Masking Area The horizontal clamping and blanking pulses of the AD9974 are fully programmable to suit a variety of applications. Individual control is provided for CLPOB, PBLK, and HBLK during the different regions of each field. This allows the dark pixel clamping and blanking patterns to be changed at each stage of the readout to accommodate different image transfer timing and high speed line shifts. Additionally, the AD9974 allows the CLPOB and PBLK signals to be disabled during certain lines in the field without changing any of the existing pattern settings. There are three sets of start and end registers for both CLPOB and PBLK that allow the creation of up to three masking areas for each signal. Individual CLPOB and PBLK Patterns The AFE horizontal timing consists of CLPOB and PBLK, as shown in Figure 23. These two signals are programmed independently using the registers in Table 11. The start polarity for the CLPOB or PBLK signal is CLPOB_POL (PBLK_POL), and the first and second toggle positions of the pulse are CLPOB_TOG1 (PBLK_TOG1) and CLPOB_TOG2 (PBLK_TOG2). Both signals are active low and need to be programmed accordingly. Two separate patterns for CLPOB and PBLK can be programmed for each H-pattern, CLPOB0, CLPOB1, PBLK0, and PBLK1. The CLPOB_PAT and PBLK_PAT field registers select which of the two patterns is used in each field. For example, to use the CLPOB masking, program the CLPOBMASKSTART and CLPOBMASKEND registers to specify the starting and ending lines in the field where the CLPOB patterns are to be ignored. Figure 24 illustrates this feature. The masking registers are not specific to a certain H-pattern; they are always active for any existing field of timing. To disable the CLPOB and PBLK masking feature, set these registers to the maximum value of 0x1FFF. Note that to disable CLPOB and PBLK masking during power-up, it is recommended that CLPOBMASKSTART (PBLKMASKSTART) be set to 8191 and CLPOBMASKEND (PBLKMASKEND) be set to 0. This prevents any accidental masking caused by different register update events. Figure 34 shows how the sequence change positions divide the readout field into different regions. By assigning a different H-pattern to each region, the CLPOB and PBLK signals can change with each change in the vertical timing. HD 2 CLPOB 1 PBLK 3 ACTIVE ACTIVE PROGRAMMABLE SETTINGS: BLANK REGION ARE ACTIVE LOW). 05955-023 1START POLARITY (CLAMP AND 2FIRST TOGGLE POSITION. 3SECOND TOGGLE POSITION. Figure 23. Clamp and Preblank Pulse Placement NO CLPOB SIGNAL FOR LINE 600 NO CLPOB SIGNAL FOR LINES 6 TO 8 VD 0 1 2 597 598 HD CLPOBMASKSTART1 = 6 CLPOBMASKEND1 = 8 CLPOBMASKSTART2 = CLPOBMASKEND2 = 600 Figure 24. CLPOB Masking Example Rev. A | Page 18 of 52 05957-024 CLPOB AD9974 Table 11. CLPOB and PBLK Pattern Registers Parameter CLPOB0_TOG1 CLPOB0_TOG2 CLPOB1_TOG1 CLPOB1_TOG2 CLPOB_POL CLPOB_PAT CLPOBMASKSTART CLPOBMASKEND PBLK0_TOG1 PBLK0_TOG2 PBLK1_TOG1 PBLK1_TOG2 PBLK_POL PBLK_PAT PBLKMASKSTART PBLKMASKEND Length (Bits) 13 13 13 13 9 9 13 13 13 13 13 13 9 9 13 13 Range 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location High/low 0 to 9 settings 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location High/low 0 to 9 settings 0 to 8191 pixel location 0 to 8191 pixel location Description First CLPOB0 toggle position within the line for each V-sequence. Second CLPOB0 toggle position within the line for each V-sequence. First CLPOB1 toggle position within the line for each V-sequence. Second CLPOB1 toggle position within the line for each V-sequence. Starting polarity of CLPOB for each V-sequence [8:0] (in field registers). CLPOB pattern selection for each V-sequence [8:0] (in field registers). CLPOB mask start position: three values available (in field registers). CLPOB mask end position: three values available (in field registers). First PBLK0 toggle position within the line for each V-sequence. Second PBLK0 toggle position within the line for each V-sequence. First PBLK1 toggle position within the line for each V-sequence. Second toggle position within the line for each V-sequence. Starting polarity of PBLK for each V-sequence [8:0] (in field registers). PBLK pattern selection for each V-sequence [8:0] (in field registers). PBLK mask start position: three values available (in field registers). PBLK mask end position: three values available (in field registers). HD HBLK HBLKTOGE2 BLANK BLANK BASIC HBLK PULSE IS GENERATED USING HBLKTOGE1 AND HBLKTOGE2 REGISTERS (HBLKALT = 0). 05955-025 HBLKTOGE1 Figure 25. Typical Horizontal Blanking Pulse Placement (HBLKMODE = 0) HD HBLK H1/H3 THE POLARITY OF H1/H3 DURING BLANKING IS PROGRAMMABLE (H2/H4 POLARITY IS SEPARATELY PROGRAMMABLE) 05955-026 H1/H3 H2/H4 Figure 26. HBLK Masking Control Rev. A | Page 19 of 52 AD9974 Individual HBLK Patterns HBLK Mode 0 Operation The HBLK programmable timing shown in Figure 25 is similar to CLPOB and PBLK; however, there is no start polarity control. Only the toggle positions designate the start and the stop positions of the blanking period. Additionally, as shown in Figure 26, there is a polarity control, HBLKMASK, for H1/H3 and H2/H4 that designates the polarity of the horizontal clock signals during the blanking period. Setting HBLKMASK_H1 low sets H1 = H3 = low and HBLKMASK_H2 high sets H2 = H4 = high during the blanking. As with the CLPOB and PBLK signals, HBLK registers are available in each H-pattern group, allowing unique blanking signals to be used with different vertical timing sequences. There are six toggle positions available for HBLK. Normally, only two of the toggle positions are used to generate the standard HBLK interval. However, the additional toggle positions can be used to generate special HBLK patterns, as shown in Figure 27. The pattern in this example uses all six toggle positions to generate two extra groups of pulses during the HBLK interval. By changing the toggle positions, different patterns are created. Separate toggle positions are available for even and odd lines. If alternation is not needed, load the same values into the registers for even (HBLKTOGE) and odd (HBLKTOGO) lines. The AD9974 supports three modes of HBLK operation. HBLK Mode 0 supports basic operation and provides some support for special HBLK patterns. HBLK Mode 1 supports pixel mixing HBLK operation. HBLK Mode 2 supports advanced HBLK operation. The following sections describe each mode. Register parameters are detailed in Table 12. HBLKTOGE2 HBLKTOGE1 HBLKTOGE4 HBLKTOGE3 HBLKTOGE6 HBLKTOGE5 HBLK H1/H3 SPECIAL H-BLANK PATTERN IS CREATED USING MULTIPLE HBLK TOGGLE POSITIONS (HBLKALT = 0). 05955-027 H2/H4 Figure 27. Generating Special HBLK Patterns Table 12. HBLK Pattern Registers Register HBLK_MODE Length (Bits) 2 Range 0 to 2 HBLK modes HBLKSTART HBLKEND HBLKLEN HBLKREP HBLKMASK_H1 HBLKMASK_H2 13 13 13 13 1 1 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixels 0 to 8191 repetitions High/low High/low Description Enables different HBLK toggle position operations. 0 = normal mode. Six toggle positions available for even and odd lines. If even/odd alternation is not needed, set toggles for even/odd the same. 1 = pixel mixing mode. In addition to six toggle positions, the HBLKSTART, HBLKEND, HBLKLEN, and HBLKREP registers can be used to generate HBLK patterns. If even/odd alternation is not needed, set toggles for even/odd the same. 2 = advanced HBLK mode. Divides HBLK interval into six different repeat areas. Uses HBLKSTARTA/B/C and RA*H*REPA/B/C registers. 3 = test mode only. Do not access. Start location for HBLK in HBLK Mode 1 and HBLK Mode 2. End location for HBLK in HBLK Mode 1 and HBLK Mode 2. HBLK length in HBLK Mode 1 and HBLK Mode 2. Number of HBLK repetitions in HBLK Mode 1 and HBLK Mode 2. Masking polarity for H1 and H3 during HBLK. Masking polarity for H2 and H4 during HBLK. Rev. A | Page 20 of 52 AD9974 Register HBLKTOGO1 HBLKTOGO2 HBLKTOGO3 HBLKTOGO4 HBLKTOGO5 HBLKTOGO6 HBLKTOGE1 HBLKTOGE2 HBLKTOGE3 HBLKTOGE4 HBLKTOGE5 HBLKTOGE6 RA0H1REPA/B/C Length (Bits) 13 13 13 13 13 13 13 13 13 13 13 13 12 Range 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 15 HCLK pulses for each A, B, and C RA1H1REPA/B/C RA2H1REPA/B/C RA3H1REPA/B/C RA4H1REPA/B/C RA5H1REPA/B/C RA0H2REPA/B/C 12 12 12 12 12 12 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 15 HCLK pulses for each A, B, and C RA1H2REPA/B/C RA2H2REPA/B/C RA3H2REPA/B/C RA4H2REPA/B/C RA5H2REPA/B/C HBLKSTARTA HBLKSTARTB HBLKSTARTC HBLKALT_PAT1 12 12 12 12 12 13 13 13 3 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 15 HCLK pulses 0 to 8191 pixel location 0 to 8191 pixel location 0 to 8191 pixel location 0 to 5 even repeat area HBLKALT_PAT2 HBLKALT_PAT3 HBLKALT_PAT4 HBLKALT_PAT5 HBLKALT_PAT6 3 3 3 3 3 0 to 5 even repeat area 0 to 5 even repeat area 0 to 5 even repeat area 0 to 5 even repeat area 0 to 5 even repeat area Description First HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1. Second HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1. Third HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1. Fourth HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1. Fifth HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1. Sixth HBLK toggle position for odd lines in HBLK Mode 0 and HBLK Mode 1. First HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1. Second HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1. Third HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1. Fourth HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1. Fifth HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1. Sixth HBLK toggle position for even lines in HBLK Mode 0 and HBLK Mode 1. HBLK Repeat Area 0. Number of H1 repetitions for HBLKSTARTA/B/C in HBLK Mode 2 for even lines; odd lines are defined using HBLKALT_PAT. [3:0] RA0H1REPA. Number of H1 pulses following HBLKSTARTA. [7:4] RA0H1REPB. Number of H1 pulses following HBLKSTARTB. [11:8] RA0H1REPC. Number of H1 pulses following HBLKSTARTC. HBLK Repeat Area 1. Number of H1 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 2. Number of H1 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 3. Number of H1 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 4. Number of H1 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 5. Number of H1 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 0. Number of H2 repetitions for HBLKSTARTA/B/C in HBLK Mode 2 for even lines; odd lines are defined using HBLKALT_PAT. [3:0] RA0H2REPA. Number of H2 pulses following HBLKSTARTA. [7:4] RA0H2REPB. Number of H2 pulses following HBLKSTARTB. [11:8] RA0H2REPC. Number of H2 pulses following HBLKSTARTC. HBLK Repeat Area 1. Number of H2 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 2. Number of H2 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 3. Number of H2 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 4. Number of H2 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area 5. Number of H2 repetitions for HBLKSTARTA/B/C. HBLK Repeat Area Start Position A for HBLK Mode 2. HBLK Repeat Area Start Position B for HBLK Mode 2. HBLK Repeat Area Start Position C for HBLK Mode 2. HBLK Mode 2, Odd Field Repeat Area 0 pattern, selected from even field. Repeat areas previously defined. HBLK Mode 2, Odd Field Repeat Area 1 pattern. HBLK Mode 2, Odd Field Repeat Area 2 pattern. HBLK Mode 2, Odd Field Repeat Area 3 pattern. HBLK Mode 2, Odd Field Repeat Area 4 pattern. HBLK Mode 2, Odd Field Repeat Area 5 pattern. Rev. A | Page 21 of 52 AD9974 HBLK Mode 1 Operation Increasing H-Clock Width During HBLK Enable multiple repeats of the HBLK signal by setting HBLK_MODE to 1. In this mode, the HBLK pattern can be generated using a different set of registers: HBLKSTART, HBLKEND, HBLKLEN, and HBLKREP, along with the six toggle positions (see Figure 28). HBLK Mode 0 and HBLK Mode 1 allow the H1 to H4 pulse width to be increased during the HBLK interval. As shown in Figure 29, the H-clock frequency can be reduced by a factor of 1/2, 1/4, 1/6, 1/8, 1/10, 1/12, and so on, up to 1/30. To enable this feature, the HCLK_WIDTH register (Address 0x34[7:4]) is set to a value between 1 and 15. When this register is set to 0, the wide HCLK feature is disabled. The reduced frequency occurs only for H1 to H4 pulses that are located within the HBLK area. Separate toggle positions are available for even and odd lines. If alternation is not needed, load the same values into the registers for even (HBLKTOGE) and odd (HBLKTOGO) lines. Generating HBLK Line Alternation HBLK Mode 0 and HBLK Mode 1 provide the ability to alternate different HBLK toggle positions on even and odd lines. Separate toggle positions are available for even and odd lines. If even/odd line alternation is not required, load the same values into the registers for even (HBLKTOGE) and odd (HBLKTOGO) lines. The HCLK_WIDTH register is generally used in conjunction with special HBLK patterns to generate vertical and horizontal mixing in the CCD. Note that the wide HCLK feature is available only in HBLK Mode 0 and HBLK Mode 1, not in HBLK Mode 2. Table 13. HCLK Width Register Register HCLK_WIDTH Length (Bits) 4 Description Controls H1 to H4 width during HBLK as a fraction of pixel rate. 0 = same frequency as pixel rate. 1 = 1/2 pixel frequency, that is, doubles the HCLK pulse width. 2 = 1/4 pixel frequency. 3 = 1/6 pixel frequency. 4 = 1/8 pixel frequency. 5 = 1/10 pixel frequency. … 15 = 1/30 pixel frequency. HBLKTOGE2 HBLKTOGE4 HBLKSTART HBLKTOGE1 HBLKTOGE3 HBLKEND HBLK HBLKLEN HBLKREP = 3 H1/H3 HBLKREP NUMBER 1 HBLKREP NUMBER 2 05955-028 H2/H4 HBLKREP NUMBER 3 H-BLANK REPEATING PATTERN IS CREATED USING HBLKLEN AND HBLKREP REGISTERS. Figure 28. HBLK Repeating Pattern Using HBLKMODE = 1 HBLK H1/H3 1/FPIX 2 × (1/FPIX) H-CLOCK FREQUENCY CAN BE REDUCED DURING HBLK BY 1/2 (AS SHOWN), 1/4, 1/6, 1/8, 1/10, 1/12, AND SO ON, UP TO 1/30 USING HBLKWIDTH REGISTER. Figure 29. Generating Wide H-Clock Pulses During HBLK Interval Rev. A | Page 22 of 52 05955-029 H2/H4 AD9974 Figure 31 shows the following example: HBLK Mode 2 Operation RA0H1REPA/RA0H1REPB/RA0H1REPC = RA0H2REPA/RA0H2REPB/RA0H2REPC = RA1H1REPA/RA1H1REPB/RA1H1REPC = RA1H2REPA/RA1H2REPB/RA1H2REPC = 2. HBLK Mode 2 allows more advanced HBLK pattern operation. If unevenly spaced HCLK pulses in multiple areas are needed, HBLK Mode 2 can be used. Using a separate set of registers, HBLK Mode 2 can divide the HBLK region into up to six repeat areas (see Table 12). As shown in Figure 31, each repeat area shares a common group of toggle positions, HBLKSTARTA, HBLKSTARTB, and HBLKSTARTC. However, the number of toggles following each start position can be unique in each repeat area by using the RAH1REP and RAH2REP registers. As shown in Figure 30, setting the RAH1REPA/RAH1REPB/ RAH1REPC or RAH2REPA/RAH2REPB/RAH2REPC registers to 0 masks HCLK groups from appearing in a particular repeat area. Figure 31 shows only two repeat areas being used, although six are available. It is possible to program a separate number of repeat area repetitions for H1 and H2, but generally the same value is used for both H1 and H2. HD CREATE UP TO 3 GROUPS OF TOGGLES A, B, C COMMON IN ALL REPEAT AREAS A Furthermore, HBLK Mode 2 allows a different HBLK pattern on even and odd lines. The HBLKSTARTA, HBLKSTARTB, and HBLKSTARTC registers, as well as the RAH1REPA/RAH1REPB/ RAH1REPC and RAH2REPA/RAH2REPB/RAH2REPC registers, define operation for the even lines. For separate control of the odd lines, the HBLKALT_PAT registers specify up to six repeat areas on the odd lines by reordering the repeat areas used for the even lines. New patterns are not available, but the order of the previously defined repeat areas on the even lines can be changed for the odd lines to accommodate advanced CCD operation. CHANGE NUMBER OF A, B, C PULSES IN ANY REPEAT AREA USING RA*H*REP* REGISTERS MASK A, B, C PULSES IN ANY REPEAT AREA BY SETTING RA*H*REP* = 0 B C H1 REPEAT AREA 0 REPEAT AREA 1 REPEAT AREA 2 REPEAT AREA 3 REPEAT AREA 4 REPEAT AREA 5 HBLKSTART HBLKEND 05955-030 H2 Figure 30. HBLK Mode 2 Operation HD HBLKLEN HBLK HBLKSTARTA ALL RA*H*REPA, B, C REGISTERS = 2, TO CREATE 2 HCLK PULSES HBLKSTARTB HBLKSTARTC H1 RA0H1REPA RA0H1REPB RA0H1REPC RA1H1REPA RA1H1REPB RA1H1REPC RA1H2REPA RA1H2REPB RA1H2REPC H2 RA0H2REPA RA0H2REPB RA0H2REPC REPEAT AREA 1 REPEAT AREA 0 HBLKREP = 2 TO CREATE 2 REPEAT AREAS Figure 31. HBLK Mode 2 Registers Rev. A | Page 23 of 52 HBLKEND 05955-031 HBLKSTART AD9974 For example, if the desired toggle position is 100, CLPOB_TOG should be set to 88 (that is, 100 − 12). Figure 49 shows the 12-cycle pipeline delay referenced to the falling edge of HD. HBLK, PBLK, and CLPOB Toggle Positions The AD9974 uses an internal horizontal pixel counter to position the HBLK, PBLK, and CLPOB toggle positions. The horizontal counter does not reset to 0 until 12 CLI periods after the falling edge of HD. This 12-cycle pipeline delay must be considered when determining the register toggle positions. For example, if CLPOB_TOG1 is 100 and the pipeline delay is not considered, the final toggle position is applied at 112. To obtain the correct toggle positions, the toggle position registers must be set to the desired toggle position minus 12. PIXEL NO. Caution Toggle positions cannot be programmed during the 12-cycle delay from the HD falling edge until the H-counter has reset. See Figure 33 for an example of this restriction. 0 60 100 103 112 HD 1 2 H1 3 1. 2. 3. 4. HBLKTOG1 HBLKTOG2 CLPOB_TOG1 CLPOB_TOG2 DESIRED TOGGLE POSITION ACTUAL REGISTER VALUE 60 100 103 112 (60 – 12) = 48 (100 – 12) = 88 (103 – 12) = 91 (112 – 12) = 100 4 05955-032 CLPOB Figure 32. Example of Register Setting to Obtain Desired Toggle Positions VD H-COUNTER RESET HD NO TOGGLE POSITIONS ALLOWED IN THIS AREA X X X X N-12 N-11 N-10 N-9 N-8 N-7 N-6 N-5 N-4 N-3 N-2 N-1 N 0 NOTES 1. TOGGLE POSITIONS CANNOT BE PROGRAMMED WITHIN 12 PIXELS OF PIXEL 0 LOCATION. Figure 33. Restriction for Toggle Position Placement Rev. A | Page 24 of 52 1 2 05955-033 H-COUNTER (PIXEL COUNTER) AD9974 COMPLETE FIELD—COMBINING H-PATTERNS H-Pattern Selection After the H-patterns are created, they combine to create different readout fields. A field consists of up to nine different regions determined by the SCP registers. Within each region, a different H-pattern group can be selected up to a maximum of 32 groups. Registers to control the H-patterns are located in the field registers. Table 31 describes the field registers. The H-patterns are stored in the HPAT memory, as described in Table 20. The user decides how many H-pattern groups are required, up to a maximum of 32, and then uses the HPAT_SEL registers to select which H-pattern group is output in each region of the field. Figure 34 shows how to use the HPAT_SEL and SCP registers. The SCP registers create the line boundaries for each region. SCP 1 SCP 0 SCP 2 SCP 3 SCP 4 SCP 5 SCP 8 VD REGION 0 REGION 1 REGION 2 REGION 3 REGION 4 REGION 8 HPAT_SEL0 HPAT_SEL1 HPAT_SEL2 HPAT_SEL3 HPAT_SEL4 HPAT_SEL8 H-PATTERNS FIELD SETTINGS: 1. SEQUENCE CHANGE POSITIONS (SCP0-8) DEFINE EACH OF THE NINE AVAILABLE REGIONS IN THE FIELD. 2. HPAT_SEL SELECTS THE DESIRED H-PATTERN FOR EACH REGION. 05955-034 HD Figure 34. Complete Field Divided into Regions Table 14. Field Registers Register SCPx HPAT_SELx CLPOB_POL CLPOB_PAT CLPOBMASKSTARTx, CLPOBMASKENDx PBLK_POL PBLK_PAT PBLKMASKSTARTx, PBLKMASKENDx, Length (Bits) 13 5 9 9 13 Range 0 to 8191 line number 0 to 31 H-patterns High/low 0 to 9 patterns Number of lines Description Sequence change position for each region. Selects an individual line. Selected H-pattern for each region of the field. CLPOB start polarity settings for each region of the field. CLPOB pattern selector for each region of the field. CLPOB mask positions for up to three masking configurations. 9 9 13 High/low 0 to 9 patterns Number of lines PBLK start polarity settings for each region of the field. PBLK pattern selector for each region of the field. PBLK mask positions for up to three masking configurations. Rev. A | Page 25 of 52 AD9974 MODE REGISTERS The mode registers contain registers to select the final field timing of the AD9974. Typically, all of the field and H-pattern group information is programmed into the AD9974 at startup. During operation, the mode registers allow the user to select any combination of field timing to meet the current requirements of the system. The advantage of using the mode registers in conjunction with preprogrammed timing is that they greatly reduce the system programming requirements during camera operation. Only a few register writes are required when the camera operating mode is changed, rather than having to write in all of the vertical timing information with each camera mode change. A basic still camera application can require five fields of horizontal timing: one for draft mode operation, one for autofocusing, and three for still image readout. With the AD9974, all of the register timing information for the five fields is loaded at startup. Then, during camera operation, the mode registers select which field timing to activate, depending on how the camera is being used. The AD9974 supports up to seven field sequences selected from up to 31 preprogrammed field groups using the FIELD_SEL registers. When FIELDNUM is greater than 1, the AD9974 starts with Field 1 and increments to each Field n at the start of each VD. Figure 35 provides examples of mode configuration settings. This example assumes to have four field groups, Field Group 0 to Field Group 3, stored in memory. Table 15. Mode Registers Register HPATNUM FIELDNUM FIELD_SEL1 FIELD_SEL2 FIELD_SEL3 FIELD_SEL4 FIELD_SEL5 FIELD_SEL6 FIELD_SEL7 Length (Bits) 5 3 5 5 5 5 5 5 5 Range 0 to 31 H-pattern groups 0 to 7 fields 0 to 31 field groups 0 to 31 field groups 0 to 31 field groups 0 to 31 field groups 0 to 31 field groups 0 to 31 field groups 0 to 31 field groups Description Total number of H-pattern groups, starting at Address 0x800. Total number of applied fields. Set to 1 for single-field operation. Selected first field. Selected second field. Selected third field. Selected fourth field. Selected fifth field. Selected sixth field. Selected seventh field. Rev. A | Page 26 of 52 AD9974 H-PATTERN MEMORY FIELD 0 FIELD 1 FIELD 2 FIELD 3 EXAMPLE 1: TOTAL FIELDS = 3, FIRST FIELD = FIELD 0, SECOND FIELD = FIELD 1, THIRD FIELD = FIELD 2 FIELD_SEL1 = 0 FIELD 0 FIELD_SEL2 = 1 FIELD 1 FIELD_SEL3 = 2 FIELD 2 EXAMPLE 2: TOTAL FIELDS = 1, FIRST FIELD = FIELD 3 FIELD_SEL1 = 3 FIELD 3 EXAMPLE 3: TOTAL FIELDS = 4, FIRST FIELD = FIELD 5, SECOND FIELD = FIELD 1, THIRD FIELD = FIELD 4, FOURTH FIELD = FIELD 2 FIELD 5 FIELD_SEL2 = 1 FIELD 1 FIELD_SEL3 = 4 FIELD 4 FIELD_SEL4 = 2 FIELD 2 05955-035 FIELD_SEL1 = 5 Figure 35. Example of Mode Configurations Rev. A | Page 27 of 52 AD9974 Figure 36 shows an example of a CCD layout. The horizontal register contains 28 dummy pixels that occur on each line clocked from the CCD. In the vertical direction, there are 10 optical black (OB) lines at the front of the readout and two at the back of the readout. The horizontal direction has four OB pixels in the front and 48 in the back. Figure 37 shows the basic sequence layout to use during the effective pixel readout. The 48 OB pixels at the end of each line are used for the CLPOB signals. PBLK is optional and is often used to blank the digital outputs during the HBLK time. HBLK is used during the vertical shift interval. The CLPOBMASK registers are also useful for disabling the CLPOB on a few lines without affecting the setup of the clamping sequences. It is important to use CLPOB only during valid OB pixels. During other portions on the frame timing, such as vertical blanking or SG line timing, the CCD does not output valid OB pixels. Any CLPOB pulse that occurs during this time causes errors in clamping operation and, therefore, changes in the black level of the image. 2 VERTICAL OB LINES Because PBLK is used to isolate the CDS input (see the Analog Front End Description and Operation section), do not use the PBLK signal during CLPOB operation. The change in the offset behavior that occurs during PBLK impacts the accuracy of the CLPOB circuitry. V EFFECTIVE IMAGE AREA 10 VERTICAL OB LINES The HBLK, CLPOB, and PBLK parameters are programmed in the V-sequence registers. More elaborate clamping schemes, such as adding a separate sequence to clamp all the shielded OB lines, can be used. This requires configuring a separate V-sequence for clocking out the OB lines. H 48 OB PIXELS 4 OB PIXELS HORIZONTAL CCD REGISTER 05955-036 HORIZONTAL TIMING SEQUENCE EXAMPLE 28 DUMMY PIXELS Figure 36. Example CCD Configuration OPTICAL BLACK OB HD CCD OUTPUT VERTICAL SHIFT DUMMY EFFECTIVE PIXELS OPTICAL BLACK VERT. SHIFT SHP SHD H1/H3 H2/H4 HBLK PBLK NOTES 1. PBLK ACTIVE (LOW) SHOULD NOT BE USED DURING CLPOB ACTIVE (LOW). Figure 37. Horizontal Sequence Example Rev. A | Page 28 of 52 05955-037 CLPOB AD9974 ANALOG FRONT END DESCRIPTION AND OPERATION 0.1µF 0.1µF REFB REFT 0.4V 1.4V AD9974 DC RESTORE SHP PBLK (WHEN DCBYP = 1) SHP SHD 0.1µF CCDIN DOUT PHASE INTERNAL VREF 2V FULL SCALE 6dB ~ 42dB S11 –3dB, 0dB, +3dB, +6dB S22 PBLK 1S1 2S2 CDS GAIN REGISTER CLI VGA GAIN REGISTER DAC PRECISION TIMING GENERATION DOUT OPTICAL BLACK CLAMP DIGITAL FILTER CLPOB 14 CLPOB PBLK IS NORMALLY CLOSED. IS NORMALLY OPEN. DOUT SHP SHD PHASE OUTPUT DATA LATCH 14-BIT ADC VGA CDS BLANK TO ZERO OR CLAMP LEVEL CLAMP LEVEL REGISTER PBLK VD V-H TIMING GENERATION HD 05955-038 1.2V Figure 38. Channel A and Channel B Analog Front End Functional Block Diagram During the PBLK active time, the ADC outputs can be programmed to output all 0s or the programmed clamp level. The AD9974 signal processing chain is shown in Figure 38. Each processing step is essential for achieving a high quality image from the raw CCD pixel data. DC Restore To reduce the large dc offset of the CCD output signal, a dc restore circuit is used with an external 0.1 μF series coupling capacitor. This restores the dc level of the CCD signal to approximately 1.2 V, making it compatible with the 1.8 V core supply voltage of the AD9974. The dc restore switch is active during the SHP sample pulse time. The dc restore circuit can be disabled when the optional PBLK signal is used to isolate large-signal swings from the CCD input (see the Analog Preblanking section). Bit 6 of Address 0x00 controls whether the dc restore is active during the PBLK interval. Analog Preblanking During certain CCD blanking or substrate clocking intervals, the CCD input signal to the AD9974 may increase in amplitude beyond the recommended input range. The PBLK signal can be used to isolate the CDS input from large-signal swings. As shown in Figure 38, when PBLK is active (low), the CDS input is isolated from the CCDIN pin (S1 open) and is internally shorted to ground (S2 closed). Note that because the CDS input is shorted during PBLK, the CLPOB pulse should not be used during the same active time as the PBLK pulse. Correlated Double Sampler (CDS) The CDS circuit samples each CCD pixel twice to extract the video information and to reject low frequency noise. The timing shown in Figure 21 illustrates how the two internally generated CDS clocks, SHP and SHD, are used to sample the reference level and data level of the CCD signal, respectively. The placement of the SHP and SHD sampling edges is determined by the setting of the SHPLOC and SHDLOC register located at Address 0x36. Placement of these two clock signals is critical for achieving the best performance from the CCD. The CDS gain is variable in four steps by using the AFE Register Address 0x04: −3 dB, 0 dB (default), +3 dB, and +6 dB. Improved noise performance results from using the +3 dB and +6 dB settings, but the input range is reduced (see Table 5). Rev. A | Page 29 of 52 AD9974 (N) SIGNAL SAMPLE Input Configurations (N) RESET SAMPLE The CDS circuit samples each CCD pixel twice to extract the video information and reject low frequency noise (see Figure 39). There are three possible configurations for the CDS: inverting CDS mode, noninverting CDS mode, and SHA mode. The CDSMODE register (Address 0x00[9:8]) selects which configuration is used. (N + 1) RESET SAMPLE VDD RESET LEVEL (VRST ) 05955-041 SHP SIGNAL LEVEL (VFS) CCDINP SHA1 Figure 41. Traditional Inverting CDS Signal DIFF AMP CCDINM CDS_OUT Table 16. Inverting Voltage Levels Signal Level Saturation Reset Supply Voltage 05955-039 SHA2 SHD Figure 39. CDS Block Diagram (Conceptual) Symbol VFS VRST VDD Min (mV) VDD − 500 1600 Typ (mV) 1000 VDD − 300 1800 Max (mV) 1400 VDD 2000 Inverting CDS Mode Noninverting Input For this configuration, the signal from the CCD is applied to the positive input of the (CCDINP) CDS system with the minus side (CCDINM) grounded (see Figure 40). The CDSMODE register setting for this configuration is 0x00. Traditional CCD applications use this configuration with the reset level established below the AVDD supply level by the AD9974 dc restore circuit, at approximately 1.5 V. The maximum saturation level is 1.0 V below the reset level, as shown in Figure 41 and Table 16. A maximum saturation voltage of 1.4 V is also possible when using the minimum CDS gain setting. If the noninverting input is desired, the reset (or black) level signal is established at a voltage above ground potential. Saturation (or white) level is approximately 1 V. Samples are taken at each signal level. See Figure 42 and Table 17. SIGNAL LEVEL (VFS) (N) RESET SAMPLE AD9974 GND (N) SIGNAL SAMPLE 05955-042 RESET LEVEL (VRST) CCDINP IMAGE SENSOR (N + 1) RESET SAMPLE Figure 42. Noninverting CDS Signal SHA/ CDS CCDINM Figure 40. Single Input CDS Configuration 05955-040 Table 17. Noninverting Voltage Levels NOTES 1. COUPLING CAPACITOR IS NOT REQUIRED FOR CERTAIN BLACK LEVEL REFERENCE VOLTAGES. Signal Level Saturation Reset Rev. A | Page 30 of 52 Symbol VFS VRST Min (mV) 0 Typ (mV) 1000 250 Max (mV) 1400 500 AD9974 SHA Mode—Differential Input Configuration Referring to Figure 46 and Table 19, the CCDINM signal is a constant dc voltage set at a level above ground potential. The sensor signal is applied to the other input, and samples are taken at the signal minimum and at a point of signal maximum. The resulting differential signal is the difference between the signal and the reference voltage. This configuration uses a differential input sample/hold amplifier (SHA) (see Figure 43). AD9974 CCDINP IMAGE SENSOR SHA/ CDS (N + 1) SIGNAL SAMPLE (N) SIGNAL SAMPLE 05955-043 CCDINM Figure 43. SHA Mode—Differential Input Configuration In this configuration, a signal is applied to the CCDINP input and, simultaneously, an inverse signal is applied to the CCDINM input. Sampling occurs on both signals at the same time. This creates the differential output for amplification and the ADC (see Figure 44 and Table 18). PEAK SIGNAL LEVEL (VFS) INPUT_POS BLACK SIGNAL LEVEL (VBLK) 05955-046 INPUT_NEG MINIMUM SIGNAL LEVEL (VMIN) GND (N + 1) SIGNAL SAMPLE Figure 46. SHA Mode—Single–Ended Input Signal (DC-Coupled) (N) SIGNAL SAMPLE Table 19. SHA Mode—Single-Ended Input Voltages Signal Level Black Saturation Minimum INPUT_POS BLACK SIGNAL LEVEL (VBLK) PEAK SIGNAL LEVEL (VFS) INPUT_NEG Figure 44. SHA Mode—Differential Input Signal Table 18. SHA Mode—Differential Voltage Levels Signal Level Black Saturation Minimum Symbol VBLK VFS VMIN Min (mV) 1000 0 Typ (mV) 0 VDD − 300 1800 Max (mV) 1400 0 Max (mV) SHA Timing Control 1400 When SHA mode is selected, only the SHPLOC setting is used to sample the input signal, but the SHDLOC signal should still be programmed to an edge setting of SHPLOC + 32. The SHA mode can also be used in a single-ended fashion, with the signal from the image sensor applied to the CDS/SHA using a single input, CCDINP. This is similar to the differential configuration, except in this case, the CCDINM line is held at a constant dc voltage, establishing a reference level that matches the image sensor reference voltage (see Figure 45). AD9974 CCDINP SHA/ CDS 05955-045 CCDINM NOTES 1. DC VOLTAGE ABOVE GROUND MAYBE USED TO MATCH THE SENSOR REFERENCE LEVEL. Typ (mV) 0 1000 The timing shown in Figure 21 illustrates how the two internally generated CDS clocks, SHP and SHD, are used to sample the reference level and data level of the CCD signal, respectively. The placement of the SHP and SHD sampling edges is determined by the setting of the SHPLOC and SHDLOC register located at Address 0x36. Placement of these two clock signals is critical in achieving the best performance from the CCD. SHA Mode—DC-Coupled, Single–Ended Input IMAGE SENSOR Min (mV) CDS Timing Control 05955-044 MINIMUM SIGNAL LEVEL (VMIN) GND Symbol VBLK VFS VMIN Figure 45. SHA Mode—Single–Ended Input Configuration, DC-Coupled Rev. A | Page 31 of 52 AD9974 Variable Gain Amplifier Optical Black Clamp The VGA stage provides a gain range of approximately 6 dB to 42 dB, programmable with 10-bit resolution through the serial digital interface. A gain of 6 dB is needed to match a 1 V input signal with the ADC full-scale range of 2 V. When compared to 1 V full-scale systems, the equivalent gain range is 0 dB to 36 dB. The optical black clamp loop is used to remove residual offsets in the signal chain and track low frequency variations in the CCD black level. During the optical black (shielded) pixel interval on each line, the ADC output is compared with a fixed black level reference, selected by the user in the clamp level register. The value can be programmed between 0 LSB and 1023 LSB in 1023 steps. The VGA gain curve follows a linear-in-dB characteristic. The exact VGA gain is calculated for any gain register value by The resulting error signal is filtered to reduce noise, and the correction value is applied to the ADC input through a DAC. Normally, the optical black clamp loop is turned on once per horizontal line, but this loop can be updated more slowly to suit a particular application. If external digital clamping is used during postprocessing, the AD9974 optical black clamping can be disabled using Bit 3 in AFE Register Address 0x00. When the loop is disabled, the clamp level register can still be used to provide fixed offset adjustment. Gain (dB) = (0.0359 × Code) + 5.1 dB where Code is the range of 0 to 1023. 42 VGA GAIN (dB) 36 30 24 Note that if the CLPOB loop is disabled, higher VGA gain settings reduce the dynamic range because the uncorrected offset in the signal path is gained up. 18 6 0 127 255 383 511 639 767 VGA GAIN REGISTER CODE 895 1023 05955-047 12 Figure 47. VGA Gain Curve ADC The AD9974 uses a high performance ADC architecture optimized for high speed and low power. Differential nonlinearity (DNL) performance is typically better than 0.5 LSB. The ADC uses a 2 V input range. See Figure 5, Figure 6, and Figure 7 for typical noise performance and linearity plots for the AD9974. The CLPOB pulse should be aligned with the optical black pixels of the CCD. It is recommended that the CLPOB pulse duration be at least 20 pixels wide. Shorter pulse widths can be used, but the ability of the loop to track low frequency variations in the black level is reduced. See the Horizontal Clamping and Blanking section for more timing information. Digital Data Outputs The AD9974 digital output data is latched using the DOUTPHASE register value, as shown in Figure 38. Output data timing is shown in Figure 22. The switching of the data outputs can couple noise back into the analog signal path. To minimize any switching noise while using default SHPLOC and SHDLOC, it is recommended that the DOUTPHASEP register be set to a value between 36 and 47. Other settings can produce good results, but experimentation is necessary. Rev. A | Page 32 of 52 AD9974 APPLICATIONS INFORMATION RECOMMENDED POWER-UP SEQUENCE 5. When the AD9974 is powered up, the following sequence is recommended (see Figure 48 for each step). 6. 1. 2. 3. 4. Turn on the power supplies for the AD9974 and apply CLI clock. There is no required sequence for turning on each supply. Although the AD9974 contains an on-chip power-on reset, a software reset of the internal registers is recommended. Write 1 to the SW_RST register (Address 0x10) to reset all the internal registers to their default values. This bit is selfclearing and automatically resets to 0. Write to the desired registers to configure high speed timing and horizontal timing. Note that all TESTMODE registers must be written as described in the Complete Register Listing section. To place the part into normal power operation, write 0 to the STANDBY and REFBUF_PWRDN registers (Address 0x00). The Precision Timing core must be reset by writing 1 to the TGCORE_RST register (Address 0x14). This starts the internal timing core operation. Write 1 to the OUT_CONTROL register (Address 0x11). The next VD/HD falling edge allows register updates to occur, including OUT_CONTROL, which enables all clock outputs. Additional Restrictions When operating, note the following restrictions: • • The HD falling edge should be located in the same CLI clock cycle as the VD falling edge or after the VD falling edge. The HD falling edge should not be located between one and five cycles prior to the VD falling edge. If possible, perform all start-up serial writes with VD and HD disabled. This prevents unknown behavior caused by partial updating of registers before all information is loaded. The internal horizontal counter is reset 12 CLI cycles after the falling edge of HD. See Figure 49 for details on how the internal counter is reset. AD9974 SUPPLIES POWER SUPPLIES 0V 1 CLI (INPUT) 2 3 4 5 6 SERIAL WRITES 1V VD (INPUT) 1ST FIELD 1H HD (INPUT) CLOCKS ACTIVE WHEN OUTCONTROL REGISTER IS UPDATED AT VD/HD EDGE H-CLOCKS H2, H4 H1, H3, RG Figure 48. Recommended Power-Up Sequence Rev. A | Page 33 of 52 05955-048 HI-Z BY DEFAULT AD9974 Example Register Settings for Power-Up The following settings can be used for basic operation. A single CLPOB pulse is used with only H-pattern and one field. Additional HPATS and FIELDS can be added, as needed, along with different CLPOB toggle positions. 010 028 800 801 802 803 804 805 806 807 808 809 80a 80b 80c 80d 80e 80f 810 811 812 813 814 815 816 817 818 819 81a 81b 81c 81d 81e 81f 02a 02b 02c 000 014 011 0000001 0000001 0064000 3ffffff 3ffffff 0064000 3ffffff 3ffffff 0000000 0000000 0000000 0000000 0000000 0000000 00dc05a 3ffffff 3ffffff 3ffffff 1000000 1000800 1000800 1000800 0000800 0000000 0000000 0000000 0000001 1000800 1000800 1000800 0000001 1000800 0000000 0000000 0000001 0000000 0000000 0000008 0000001 0000001 //software reset //total number of H-Pattern groups = 1 //HPAT0 HBLKTOGO1, TOGO2 settings //unused HBLK odd toggles set to zero or max value //unused HBLK odd toggles set to zero or max value //HPAT0 HBLKTOGE1, TOGE2 settings //unused HBLK Even toggles set to zero or max value //unused HBLK Even toggles set to zero or max value //HBLK StartA, B are not used //HBLK StartC is not used //HBLK alternation patterns are not used //HBLKLEN, HBLKREP not used, HBLK masking pol = 0 //HBLKSTART, end not used //test, set to zero //CLPOB pat 0 toggles //CLPOB pat 1 toggles not used, set to max //PBLK pat 0 toggles not used, set to max //PBLK pat 1 toggles not used, set to max //FIELD0 SCP0, SCP1 //SCP2, SCP3 set same as SCP1 //SCP4, SCP5 set same as SCP1 //SCP6, SCP7 set same as SCP1 //SCP8 set same as SCP1 //select HPAT0 for all regions //select HPAT0 for all regions //test, set to zero //CLPOB start polarity = HIGH //CLPOB masking set to highest SCP value (no mask) //CLPOB masking set to highest SCP value (no mask) //CLPOB masking set to highest SCP value (no mask) //PBLK start polarity = HIGH //PBLK masking set to highest SCP value (no mask) //PBLK masking set to highest SCP value (no mask) //PBLK masking set to highest SCP value (no mask) //total number of fields = 1 //field select = FIELD0 //field select = FIELD0 //AFE settings //reset TGCORE //enable outputs Rev. A | Page 34 of 52 AD9974 VD tVDHD HD 3ns MIN CLI 3ns MIN tCLIDLY SHD INTERNAL HD INTERNAL H-COUNTER RESET 11.5 CYCLES X X X X X X X X X X X X X X 0 1 2 NOTES 1. EXTERNAL HD FALLING EDGE IS LATCHED BY CLI RISING EDGE, THEN LATCHED AGAIN BY SHD INTERNAL FALLING EDGE. 2. INTERNAL H-COUNTER IS ALWAYS RESET 11.5 CLOCK CYCLES AFTER THE INTERNAL HD FALLING EDGE. 3. DEPENDING ON THE VALUE OF SHDLOC, H-COUNTER RESET CAN OCCUR 12 OR 13 CLI CLOCK EDGES AFTER THE EXTERNAL HD FALLING EDGE. 4. SHPLOC = 0 IS SHOWN IN THE ABOVE EXAMPLE. IN THIS CASE, THE H-COUNTER RESET OCCURS 12 CLI RISING EDGES AFTER HD FALLING EDGE. 5. HD FALLING EDGE SHOULD OCCUR COINCIDENT WITH VD FALLING EDGE (WITHIN SAME CLI CYCLE) OR AFTER VD FALLING EDGE. HD FALLING EDGE SHOULD NOT OCCUR WITHIN 1 AND 5 CLI CYCLES IMMEDIATELY BEFORE VD FALLING EDGE. 05955-049 H-COUNTER (PIXEL COUNTER) Figure 49. Horizontal Counter Pipeline Delay VD H-COUNTER RESET HD NO TOGGLE POSITIONS ALLOWED IN THIS AREA X X X X X X N – 11 N – 10 N – 9 N – 8 N – 7 N – 6 N – 5 N – 4 N – 3 N – 2 N – 1 NOTES 1. TOGGLE POSITIONS CANNOT BE PROGRAMMED WITHIN 12 PIXELS OF PIXEL 0 LOCATION. N 0 1 2 05955-050 H-COUNTER (PIXEL COUNTER) Figure 50. No-Toggle Positions Additional Restrictions • When operating, note the following restrictions: • The HD falling edge should be located in the same CLI clock cycle as the VD falling edge or later than the VD falling edge. The HD falling edge should not be located within 1 cycle prior to the VD falling edge. If possible, perform all start-up serial writes with VD and HD disabled. This prevents unknown behavior caused by partial updating of registers before all information is loaded. The internal horizontal counter is reset 12 CLI cycles after the falling edge of HD. See Figure 49 for details on how the internal counter is reset. Rev. A | Page 35 of 52 AD9974 STANDBY MODE OPERATION The AD9974 contains two standby modes to optimize the overall power dissipation in a particular application. Bit 1 and Bit 0 of Address 0x00 control the power-down state of the device. STANDBY[1:0] = 00 = normal operation (full power) STANDBY[1:0] = 01 = reference standby mode STANDBY[1:0] = 10 or 11 = total shutdown mode (lowest power) When returning from total shutdown mode to normal operation, the timing core must be reset at least 100 μs after the STANDBY register is written to. There is an additional register to disable the internal voltage reference buffer (Address 0x00[2]) independently. By default the buffer is disabled, but it must be enabled for normal operation. CLI FREQUENCY CHANGE Table 20 summarizes the operation of each power-down mode. The OUT_CONTROL register takes priority over the reference standby mode in determining the digital output states, but total shutdown mode takes priority over OUT_CONTROL. Total shutdown mode has the lowest power consumption. If the input clock, CLI, is interrupted or changes to a different frequency, the timing core must be reset for proper operation. After the CLI clock has settled to the new frequency, or the previous frequency has resumed, write 0 and then 1 to the TGCORE_RST register (Address 0x14). This guarantees proper timing core operation. Table 20. Standby Mode Operation I/O Block AFE Timing Core H1 H2 H3 H4 HL RG DOUT Total Shutdown (Default) 1, 2 Off Off High-Z High-Z High-Z High-Z High-Z High-Z Low 3 OUT_CONTROL = Low2 No change No change Low High Low High Low Low Low 1 Reference Standby Only REFT, REFB on On Low (4.3 mA) High (4.3 mA) Low (4.3 mA) High (4.3 mA) Low (4.3 mA) Low (4.3 mA) Low To exit total shutdown, write 00 to STANDBY (Address 0x00, Bits[1:0]), then reset the timing core after 100 μs to guarantee proper settling. Total shutdown mode takes priority over OUT_CONTROL for determining the output polarities. 3 The status of the DOUT pins is unknown at power-up. Low status is guaranteed in total shutdown mode after the power-up sequence is completed. 2 Rev. A | Page 36 of 52 AD9974 CIRCUIT CONFIGURATION The AD9974 recommended circuit configuration is shown in Figure 51. Achieving good image quality from the AD9974 requires careful attention to PCB layout. All signals should be routed to maintain low noise performance. The CCD_A and CCD_B output signals should be directly routed to Pin A1 and Pin A7, respectively, through a 0.1 μF capacitor. The master clock, CLI_X, should be carefully routed to Pin A3 and Pin A9 to minimize interference with the CCDIN_X, REFT_X, and REFB_X signals. The digital outputs and clock inputs should be connected to the digital ASIC away from the analog and CCD clock signals. Placing series resistors close to the digital output pins may help reduce digital code transition noise. If the digital outputs must drive a load larger than 20 pF, buffering is recommended to minimize additional noise. If the digital ASIC can accept gray code, the outputs of the AD9974 can be selected to output data in gray code format using Register 0x01[2]. Compared with binary coding, gray coding helps reduce potential digital transition noise. The H1_X to H4_X and RG_X traces should have low inductance to avoid excessive distortion of the signals. Heavier traces are recommended because of the large transient current demand on H1_X to H4_X from the capacitive load of the CCD. If possible, physically locating the AD9974 closer to the CCD reduces the inductance on these lines. As always, the routing path should be as direct as possible from the AD9974 to the CCD. The CLI_X and CCDIN_X PCB traces should be carefully matched in length and impedance to achieve optimal channel-to-channel matching performance. 3 V System Compatibility GROUNDING AND DECOUPLING RECOMMENDATIONS As shown in Figure 51, a single ground plane is recommended for the AD9974. This ground plane needs to be as continuous as possible, particularly around the P-type, AI-type, and A-type pins to ensure that all analog decoupling capacitors provide the lowest possible impedance path between the power and bypass pins and their respective ground pins. All high frequency decoupling capacitors need to be located as close as possible to the package pins. All the supply pins must be decoupled to ground with good quality, high frequency chip capacitors. There also needs to be a 4.7 μF or larger bypass capacitor for each main supply, that is, AVDD, RGVDD, HVDD, and DRVDD, although this is not necessary for each individual pin. In most applications, it is easier to share the supply for RGVDD and HVDD, which can be done as long as the individual supply pins are separately bypassed. A separate 3 V supply can be used for DRVDD, but this supply pin still needs to be decoupled to the same ground plane as the rest of the chip. A separate ground for DRVSS is not recommended. The reference bypass pins (REFT, REFB) must be decoupled to ground as close as possible to their respective pins. The bridge capacitor between REFT and REFB is recommended for pixel rates greater than 40 MHz. The analog input capacitor (CCDINM, CCDINP) also needs to be located close to the pin. The GND connections should be tied to the lowest impedance ground plane on the PCB. Performance does not degrade if several of these GND connections are left unconnected for routing purposes. The AD9974 typical circuit connections for a 3 V system are shown in Figure 51. This application uses an external 3.3 V supply connected to the IOVDD input of the AD0074, which also serves as the LDO input. The LDO generates a 1.8 V output for the AD9974 core supply voltages, AVDD and DVDD. The LDOOUT pin can then be connected directly to the AVDD and DVDD pins. In this configuration, the LDOEN pin is tied high to enable the LDO. Alternatively, a separate 1.8 V regulated supply voltage may be used to power the AVDD and DVDD pins. In this case, the LDOOUT pin needs to be left floating, and the LDOEN pin needs to be grounded. A typical circuit configuration for a 1.8 V system is shown in Figure 51. Rev. A | Page 37 of 52 AD9974 3V IOVDD SUPPLY 1.8V LDO OUTPUT A RG_A OUTPUT RG_A DRIVER SUPPLY A8 CCDINP_B AVDD_B A7 RGVDD_B CCDINM_B A6 C9 C10 RGVSS_B 0.1µF 0.1µF RG_B A10 B10 GND GND LDO_OUT_B B9 B6 C6 GND GND GND D6 E6 E5 GND GND D5 C5 IOVDD_B 0.1µF 0.1µF 0.1µF IOVDD_A GND B5 A5 B3 B4 LDO_OUT_A 0.1µF RG_A RGVSS_A RGVDD_A C4 A1 C3 CCDINM_A 0.1µF D9 F7 G1 F8 H1 G7 J1 H7 K1 J7 G2 K7 H2 G8 K2 H8 3V DRIVER 3V DRIVER Figure 51. Recommended Circuit Configuration Rev. A | Page 38 of 52 CLI_B COMMON MASTER CLOCK INPUT REFB_B REFT_B 0.1µF 0.1µF 0.1µF SL_B SDATA_B SCK_B HD_B VD_B 3 HD COMMON INPUT VD COMMON INPUT HVDD_B HVSS_B 0.1µF H1B 4 H2B SERIAL INTERFACE FOR CHN B H1B–H4B DRIVER SUPPLY H1B–H4B OUTPUTS H3B H4B DVDD_B DVSS_B 1.8V ANALOG 0.1µF SUPPLY D0_B (LSB) D1_B D2_B D3_B D4_B D5_B D6_B D7_B H9 J9 AVSS_B D8_B D9_B K10 D10_B J10 H10 D11_B D12_B G10 K9 D13_B (MSB) DRVDD_B J8 DRVSS_B F6 G6 GND GND GND GND D10_A H6 G9 J6 K8 H3 K6 G3 0.1µF 14 D10 F2 D9_A CHN A DATA OUTPUTS CCDINP_A NOT DRAWN TO SCALE F9 F1 J3 D8_A D3 GND D7_A AD9974 K5 D6_A F10 D4 J5 D5_A F3 GND D4_A E9 GND D3_A F4 H5 D2_A E10 GND D1_A E8 E3 G5 D0_A (LSB) E4 F5 DVSS_A E7 J2 0.1µF D8 E2 GND DVDD_A 1.8V ANALOG SUPPLY E1 GND H4A C8 DRVSS_A H3A B8 D2 0.1µF HVDD_A 0.1µF HVSS_A H1A H1A–H4A 4 OUTPUTS H2A 4.7µF C2 4.7µF + VD_A C7 K3 HD_A B2 DRVDD_A H1A–H4A DRIVER SUPPLY HD COMMON INPUT VD COMMON INPUT C1 G4 SCK_A D7 H4 SDATA_A 1.8V ANALOG SUPPLY 0.1µF A9 D12_A SL_A CCD SIGNAL_B PLUS B7 D13_A (MSB) 0.1µF SERIAL INTERFACE FOR CHN A 1.8V LDO OUTPUT B RG_B OUTPUT RG_B DRIVER SUPPLY D1 J4 0.1µF REFT_A A4 D11_A 0.1µF A2 AVSS_A REFB_A B1 CLI_A COMMON MASTER CLOCK INPUT AVDD_A + A3 4.7µF K4 CCD SIGNAL_A PLUS 1.8V ANALOG SUPPLY 3V IOVDD SUPPLY + 4.7µF 14 CHN B DATA OUTPUTS AD9974 3-WIRE SERIAL INTERFACE TIMING All of the internal registers of the AD9974 are accessed through a 3-wire serial interface. Each register consists of a 12-bit address and a 28-bit data-word. Both the 12-bit address and 28-bit dataword are written starting with the LSB. To write to each register, a 40-bit operation is required, as shown in Figure 52. Although many registers are fewer than 28 bits wide, all 28 bits must be written for each register. For example, if the register is only 20 bits wide, the upper eight bits are don’t cares and must be filled with 0s during the serial write operation. If fewer than 28 data bits are written, the register is not updated with new data. Figure 53 shows a more efficient way to write to the registers, using the AD9974 address auto-increment capability. Using this method, the lowest desired address is written first, followed by multiple 28-bit data-words. Each new 28-bit data-word is automatically written to the next highest register address. By eliminating the need to write each 12-bit address, faster register loading is achieved. Continuous write operations can be used, starting with any register location. 12-BIT ADDRESS SDATA A0 A1 A2 A3 A4 A5 A6 A7 tDS SCK 1 2 3 4 5 A8 28-BIT DATA A9 A10 A11 D0 D1 D2 D3 D25 D26 D27 tDH 6 7 8 9 10 11 12 13 14 15 16 38 tLS 39 40 tLH SL 05955-052 NOTES 1. SDATA BITS ARE LATCHED ON SCK RISING EDGES. SCK CAN IDLE HIGH OR LOW BETWEEN WRITE OPERATIONS. 2. ALL 40 BITS MUST BE WRITTEN: 12 BITS FOR ADDRESS AND 28 BITS FOR DATA. 3. IF THE REGISTER LENGTH IS <28 BITS, THEN ZEROS MUST BE USED TO COMPLETE THE 28-BIT DATA LENGTH. 4. NEW DATA VALUES ARE UPDATED IN THE SPECIFIED REGISTER LOCATION AT DIFFERENT TIMES, DEPENDING ON THE PARTICULAR REGISTER WRITTEN TO. SEE THE UPDATING OF NEW REGISTER VALUES SECTION FOR MORE INFORMATION. Figure 52. Serial Write Operation DATA FOR STARTING REGISTER ADDRESS SDATA SCK A0 1 A1 2 A2 3 A3 4 A10 11 A11 12 D0 13 D1 14 D26 39 DATA FOR NEXT REGISTER ADDRESS D27 40 D0 D1 41 42 D26 D27 67 68 D0 69 D1 70 D2 71 NOTES 1. MULTIPLE SEQUENTIAL REGISTERS CAN BE LOADED CONTINUOUSLY. 2. THE FIRST (LOWEST ADDRESS) REGISTER ADDRESS IS WRITTEN, FOLLOWED BY MULTIPLE 28-BIT DATA-WORDS. 3. THE ADDRESS AUTOMATICALLY INCREMENTS WITH EACH 28-BIT DATA-WORD (ALL 28 BITS MUST BE WRITTEN). 4. SL IS HELD LOW UNTIL THE LAST DESIRED REGISTER HAS BEEN LOADED. Figure 53. Continuous Serial Write Operation Rev. A | Page 39 of 52 05955-053 SL AD9974 LAYOUT OF INTERNAL REGISTERS The AD9974 address space is divided into two register areas, as shown in Figure 54. In the first area, Address 0x00 to Address 0x72 contain the registers for the AFE, miscellaneous functions, VD/HD parameters, I/O control, mode control, timing core, and update control functions. The second area of the address space, beginning at Address 0x800, consists of the registers for the H-pattern groups and fields. This is a configurable set of register spaces; the user can decide how many H-patterns and fields are used in a particular design. The AD9974 supports the use of up to 32 H-patterns. Register 0x28 specifies the total number of H-pattern groups. The starting address for the H-pattern groups is always 0x800. The starting address for the field registers is determined by the number of H-pattern groups. Each H-pattern group and field occupies 16 register addresses. The starting address for the field registers is based on the number of H-pattern groups and is equal to 0x800 plus the number of H-pattern groups times 16. It is important to note that the H-pattern and field registers must always occupy a continuous block of addresses. Figure 55 shows an example when three H-pattern groups and two fields are used. The starting address for the H-pattern groups is always 0x800. Because HPATNUM is 3, the H-pattern groups occupy 48 address locations (that is, 16 registers × 3 H-pattern groups). The starting address of the field registers for this example is 0x830 (that is, 0x800 + 48 (decimal)). Note that the decimal value must be converted to a hex number before adding it to 0x800. The AD9974 address space contains many unused addresses. Any undefined addresses between Address 0x00 and Address 0x7FF should not be written to; otherwise, the AD9974 may operate incorrectly. Continuous register writes should be performed carefully so that undefined registers are not written to. CONFIGURABLE REGISTER AREA FIXED REGISTER AREA ADDR 0x000 HPAT START 0x800 AFE REGISTERS MISCELLANEOUS FUNCTION REGISTERS VD/HD REGISTERS H-PATTERN GROUPS I/O REGISTERS MODE CONTROL REGISTERS TIMING CORE REGISTERS FIELD START TEST REGISTERS FIELDS UPDATE CONTROL REGISTERS INVALID—DO NOT ACCESS MAX 0xFFF NOTES 1. THE H-PATTERN AND FIELD REGISTERS MUST ALWAYS OCCUPY A CONTINUOUS BLOCK OF ADDRESSES. Figure 54. Layout of AD9974 Registers ADDR 0x800 3 H-PATTERN GROUPS (16 × 3 = 48 REGISTERS) ADDR 0x830 2 FIELDS (16 × 2 = 32 REGISTERS) UNUSED MEMORY MAX 0xFFF Figure 55. Example Register Configuration Rev. A | Page 40 of 52 05955-055 ADDR 0x850 05955-054 ADDR 0x7FF AD9974 • UPDATING OF REGISTER VALUES The internal registers of the AD9974 are updated at different times, depending on the particular register. Table 21 summarizes the three types of register updates. The tables in the Complete Register Listing section also contain a column with update type to identify when each register is updated. • SCK Updated—Some of the registers are updated immediately as the 28th data bit (D27) is written. These registers are used for functions that do not require gating with the next VD boundary, such as power-up and reset functions. • VD Updated—Many of the registers are updated at the next VD falling edge. By updating these values at the next VD edge, the current field is not corrupted, and the new register values are applied to the next field. The VD update can be further delayed past the VD falling edge by using UPDATE Register Address 0x17. This delays the VD-updated register updates to any HD line in the field. Note that the field registers are not affected by the UPDATE register. SCP Updated—All of the H-pattern group registers are updated at the next SCP when they are used. Table 21. Register Update Types Update Type SCK Updated VD Updated SCP Updated Description Register is immediately updated when the 28th data bit (D27) is clocked in. Register is updated at the VD falling edge. VD-updated registers can be delayed further by using the UPDATE register at Address 0x17. Field registers are not affected by the UPDATE register. Register is updated at the next SCP when the register is used. Rev. A | Page 41 of 52 AD9974 COMPLETE REGISTER LISTING All addresses and default values are expressed in hexadecimal. When an address contains less than 28 data bits, all remaining bits must be written as 0s. All TESTMODE registers must be set to the specified values. Table 22. AFE Registers Address 0x00 0x01 0x02 0x03 0x04 0x05 0x06 0x07 0x08 0x09 0x0A 0x0B Data Bit Content [1:0] Default Value 3 [2] 1 REFBUF_PWRDN [3] 1 CLAMPENABLE [5:4] [6] 0 0 TESTMODE PBLK_LVL [7] 0 DCBYP [9:8] 0 CDSMODE [16:10] [27:17] [1:0] [2] 0 TESTMODE Unused TESTMODE GRAYENCODE [3] [4] [27:5] [0] [27:1] [23:0] [27:24] [1:0] [27:2] [9:0] [27:10] [9:0] [27:10] [27:0] [27:0] [27:0] [27:0] [27:0] 0 0 Update SCK SCK 0 1 0 SCK FFFFFF SCK 1 VD F VD 1EC VD 0 0 0 0 0 Name STANDBY TESTMODE TESTMODE Unused TESTMODE Unused TESTMODE Unused CDSGAIN Unused VGAGAIN Unused CLAMPLEVEL Unused TESTMODE TESTMODE TESTMODE TESTMODE TESTMODE Description Standby Modes. 0 = normal operation. 1= band gap reference in standby. 2, 3 = total power-down. Reference Buffer for REFT and REFB Power Control. 0 = REFT/REFB internally driven. 1 = REFT/REFB not driven. Clamp Enable Control. 0 = disable black clamp. 1 = enable black clamp. Test Operation Only. Set to 0. PBLK Level Control. 0 = blank to 0. 1 = blank to clamp level. DC Restore Circuit Control. 0 = enable dc restore circuit during PBLK. 1 = bypass dc restore circuit during PBLK. CDS Operation. 0 = normal (inverting) CDS mode. 1 = sample and hold (SHA) mode. 2 = positive CDS mode. 3 = invalid, do not use. Test Operation Only. Set to 0. Set unused bits to 0. Test Operation Only. Set to 0. Gray Coding ADC Outputs. 0 = disable. 1 = enable. Test Operation Only. Set to 0. Test Operation Only. Set to 0. Set unused bits to 0. Test Operation Only. Set to 0. Set unused bits to 0. Test Operation Only. Set to FFFFFF. Set unused bits to 0. CDS Gain Setting. 0 = −3 dB. 1 = 0 dB (default). 2 = +3 dB. 3 = +6 dB. Set unused bits to 0. VGA Gain. 6 dB to 42 dB (0.035 dB per step). Set unused bits to 0. Optical Black Clamp Level. 0 LSB to 1023 LSB (1 LSB per step). Set unused registers to 0. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Rev. A | Page 42 of 52 AD9974 Address 0x0C 0x0D Data Bit Content [27:0] [0] Default Value 0 0 0 0x0E 0x0F [3:1] [27:4] [27:0] [27:0] Update VD VD Name TESTMODE CLIDIVIDE SCK SCK TESTMODE Unused Unused Unused Description Test Operation Only. Set to 0 if this register is accessed. CLI Divide. 1 = divide CLI input frequency by 2. Test Operation Only. Set to 0. Set unused bits to 0. Set unused register to 0 if accessed. Set unused register to 0 if accessed. Table 23. Miscellaneous Registers Address 0x10 Data Bit Content [0] Default Value 0 Update SCK Name SW_RST 0x11 [27:1] [0] 0 VD Unused OUT_CONTROL 0 SCK 0 SCK 0 SCK Unused TESTMODE Unused TESTMODE Unused TGCORE_RST 0 SCK Unused CLI_BIAS 0 SCK 0 SCK 0x14 [27:1] [1:0] [27:2] [0] [27:1] [0] 0x15 [27:1] [0] 0x12 0x13 0x16 0x17 0x18 0x19 0x1A to 0x1F [27:1] [0] [27:1] [12:0] Unused TESTMODE Unused UPDATE [13] 0 PREVENTUP [27:14] [27:0] [27:0] [27:0] 0 0 Unused TESTMODE TESTMODE Unused Description Software Reset. Bit self-clears to 0 when a reset occurs. 1 = reset Address 0x00 to Address 0xFF to default values. Set unused bits to 0. Output Control. 0 = make all outputs dc inactive. 1 = enable outputs at next VD edge. Set unused bits to 0. Test Operation Only. Set to 0. Set unused bits to 0. Test Operation Only. Set to 0. Set unused bits to 0. Timing Core Reset Bar. 0 = hold in reset. 1 = resume operation. Set unused bits to 0. Enable bias for CLI input (see Figure 11). 0 = disable bias (CLI input is dc-coupled). 1 = enable bias (CLI input is ac-coupled). Set unused bits to 0. Test Operation Only. Set to 0. Set unused bits to 0. Serial Interface Update Line. Sets the line (HD) within the field to update the VD-updated registers. Disabled when PREVENTUP = 1. Prevents normal update of VD-updated registers. 0 = normal update at VD. 1 = prevent update of VD-updated registers. Set unused bits to 0. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Set unused bits to 0. Table 24. VD/HD Registers 0x21 Data Bit Content [0] [27:1] [0] 0x22 [2:1] [27:3] [27:0] Address 0x20 Default Value 0 Update SCK 0 SCK 0 0 Name TESTMODE Unused VDHDPOL TESTMODE Unused TESTMODE Description Test Operation Only. Set to 0. Set unused bits to 0. VD/HD Active Polarity. 0 = active low. 1 = active high. Test Operation Only. Set to 0. Set unused bits to 0. Test Operation Only. Set to 0 if this register is accessed. Rev. A | Page 43 of 52 AD9974 Table 25. I/O Control Registers Address 0x23 0x24 0x25 0x26 0x27 Data Bit Content [0] [1] [2] Default Value 0 0 0 [3] [4] [7:5] [27:8] [27:0] [27:0] [27:0] [27:0] 0 0 1 Update SCK 0 0 0 0 Name TESTMODE TESTMODE IO_NVR DATA_NVR TESTMODE HCLKMODE Unused TESTMODE TESTMODE TESTMODE TESTMODE Description Test Operation Only. Set to 0. Test Operation Only. Set to 0. IOVDD Voltage Range for VD, HD, SCK, SDATA, and SL. 0 = 1.8 V. 1 = 3.3 V. The I/Os are 3 V tolerant, so there is no problem having higher than 1.8 V inputs at start-up, but this register should be set to 1 at initialization if using higher than 1.8 V supplies. DRVDD Voltage Range. Test Operation Only. Set to 0. Selects HCLK output configuration (see Table 9). Set unused bits to 0. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Test Operation Only. Set to 0 if this register is accessed. Name HPATNUM Unused Unused FIELDNUM Unused FIELD_SEL1 FIELD_SEL2 FIELD_SEL3 FIELD_SEL4 FIELD_SEL5 Unused FIELD_SEL6 FIELD_SEL7 Unused Unused Unused Unused Description Total Number of H-Pattern Groups. Set unused bits to 0. Set unused register to 0 if accessed. Total Number of Fields. Set to 1 for single-field operation. Set unused bits to 0. Selected First Field. Selected Second Field. Selected Third Field. Selected Fourth Field. Selected Fifth Field. Set unused bits to 0. Selected Sixth Field. Selected Seventh Field. Set unused bits to 0. Set unused register to 0 if this register is accessed. Set unused register to 0 if this register is accessed. Set unused register to 0 if this register is accessed. Table 26. Mode Registers Address 0x28 0x29 0x2A 0x2B 0x2C 0x2D 0x2E 0x2F Data Bit Content [4:0] [27:5] [27:0] [2:0] [27:3] [4:0] [9:5] [14:10] [19:15] [24:20] [27:25] [4:0] [9:5] [27:10] [27:0] [27:0] [27:0] Default Value 0 Update VD 0 VD 0 0 0 0 0 VD VD 0 0 VD SCK SCK SCK Rev. A | Page 44 of 52 AD9974 Table 27. Timing Core Registers Address 0x30 0x31 0x32 0x33 0x34 0x35 Data Bit Content [5:0] [7:6] [13:8] [15:14] [16] [27:17] [5:0] [7:6] [13:8] [15:14] [16] [27:17] [5:0] [7:6] [13:8] [15:14] [16] [27:17] [5:0] [7:6] [13:8] [15:14] [16] Default Value 0 Update SCK 20 0 1 0 SCK 20 0 1 0 SCK 20 0 1 0 SCK 10 0 1 Name H1POSLOC Unused H1NEGLOC TESTMODE H1POL Unused H2POSLOC Unused H2NEGLOC TESTMODE H2POL Unused TESTMODE Unused TESTMODE TESTMODE TESTMODE Unused RGPOSLOC Unused RGNEGLOC TESTMODE RGPOL [27:17] [0] 0 [1] [2] [3] [7:4] 0 0 0 0 H2BLKRETIME TESTMODE TESTMODE HCLK_WIDTH 1 Unused H1DRV [27:8] [2:0] [3] [6:4] [7] 1 SCK SCK Unused H1BLKRETIME Unused H2DRV Unused Description H1 Rising Edge Location. Set unused bits to 0. H1 Falling Edge Location. Test Operation Only. Set to 0. H1 Polarity Control. 0 = inverse of Figure 21. 1 = no inversion. Set unused bits to 0. H2 Rising Edge Location. Set unused bits to 0. H2 Falling Edge Location. Test Operation Only. Set to 0. H2 Polarity Control. 0 = inverse of Figure 21. 1 = no inversion. Set unused bits to 0. Test Operation Only. Set to 0. Set unused bits to 0. Test Operation Only. Set to 20. Test Operation Only. Set to 0. Test Operation Only. Set to 1. Set unused bits to 0. RG Rising Edge Location. Set unused bits to 0. RG Falling Edge Location. Test Operation Only. Set to 0. RG Polarity Control. 0 = inverse of Figure 21. 1 = no inversion. Set unused bits to 0. Retime H1 HBLK to Internal Clock. 0 = no retime. 1 = enable retime. Recommended setting is enable retime. Enabling retime adds one cycle delay to programmed HBLK positions. Retime H2 HBLK to Internal Clock. Test Operation Only. Set to 0 Test Operation Only. Set to 0 Enables wide H-clocks during HBLK interval. 0 = disable (see Table 13). Set unused bits to 0. H1 Drive Strength. 0 = off. 1 = 4.3 mA. 2 = 8.6 mA. 3 = 12.9 mA. 4 = 17.2 mA. 5 = 21.5 mA. 6 = 25.8 mA. 7 = 30.1 mA. Set unused bits to 0. H2 Drive Strength. Set unused bits to 0. Rev. A | Page 45 of 52 AD9974 Address 0x36 0x37 0x38 0x39 0x3A 0x3B 0x3C 0x3D Data Bit Content [10:8] [11] [14:12] [15] [18:16] [19] [22:20] [27:23] [5:0] [11:6] [17:12] [27:18] [5:0] [11:6] Default Value 1 Update Unused Unused Unused Unused Unused Unused Unused Description H3 Drive Strength. Set unused bits to 0. H4 Drive Strength. Set unused bits to 0. Test Operation Only. Set to 0. Set unused bits to 0. RG Drive Strength. Set unused bits to 0. SHD Sampling Edge Location. SHP Sampling Edge Location. SHP Width. Controls input dc restore switch active time. Set unused bits to 0. DOUT Positive Edge Phase Control. DOUT Negative Edge Phase Control. Set DOUTPHASEN = DOUTPHASEP + 0x20. 0 = DCLK tracks DOUT phase. 1 = DCLK is CLI post-Schmitt trigger and post-divider when CLIDIVIDE = 1. Data Output Clock Selection. 0 = no delay. 1 = ~4 ns. 2 = ~8 ns. 3 = ~12 ns. 0 = no invert. 1 = invert DCLK to output. Set unused bits to 0. Set unused register to 0 if this register is accessed. Set unused register to 0 if this register is accessed. Set unused register to 0 if this register is accessed. Set unused register to 0 if this register is accessed. Set unused register to 0 if this register is accessed. Set unused register to 0 if this register is accessed. SCK Name TESTMODE Unused Unused TESTMODE TESTMODE Unused Unused Description Test Operation Only. Set to 4B020. Set unused bits to 0. Set unused register to 0 if these registers are accessed. Test Operation Only. Set to F if accessed. Test Operation Only. Set to 0. Set unused bits to 0. Set unused register to 0 if these registers are accessed. SCK Unused Set unused register to 0 if these registers are accessed. 1 1 0 0 20 10 0 20 SCK SCK Name H3DRV Unused H4DRV Unused TESTMODE Unused RGDRV Unused SHDLOC SHPLOC SHPWIDTH Unused DOUTPHASEP DOUTPHASEN [12] 0 DCLKMODE [14:13] 2 CLKDATA_SEL [15] 0 INV_DCLK [27:16] [27:0] [27:0] [27:0] [27:0] [27:0] [27:0] Table 28. Test Registers—Do Not Access Address 0x3E 0x3F 0x40 0x41 to 0x4F 0x50 to 0x5F Data Bit Content [18:0] [27:19] [27:0] [3:0] [9:4] [27:10] [27:0] [27:0] Default Value 4B020 F 0 Update SCK SCK SCK Rev. A | Page 46 of 52 AD9974 Table 29. Update Control Registers Address 0x60 Data Bit Content [15:0] 0x61 [27:16] [15:0] 0x62 [27:16] [15:0] 0x63 [27:16] [15:0] 0x66 [27:16] [15:0] [27:16] [15:0] [27:16] [15:0] 0x67 [27:16] [15:0] 0x64 0x65 0x68 to 0x72 Default Value 1803 Update SCK Name AFE_UPDT_SCK E7FC SCK Unused AFE_UPDT_VD F8FD SCK Unused MISC_UPDT_SCK 0702 SCK Unused MISC_UPDT_VD FFF9 SCK 0006 SCK FFFF SCK Unused VDHD_UPDT_SCK Unused VDHD_UPDT_VD Unused TGCORE_UPDT_SCK 0000 SCK Unused TGCORE_UPDT_VD SCK Unused Unused [27:16] [27:0] Description Enable SCK update of AFE registers. Each bit corresponds to one address location. AFE_UPDT_SCK[0] = 1, update Address 0x00 on SL rising edge. AFE_UPDT_SCK[1] = 1, update Address 0x01 on SL rising edge. … AFE_UPDT_SCK[15] = 1, update Address 0x0F on SL rising edge. Set unused register = 0 if accessed. Enable VD update of AFE registers. Each bit corresponds to one address location. AFE_UPDT_VD[0] = 1, update Address 0x00 on VD rising edge. AFE_UPDT_VD[1] = 1, update Address 0x01 on VD rising edge. … AFE_UPDT_VD[15] = 1, update Address 0x0F on VD rising edge. Set unused register to 0 if accessed. Enable SCK update of miscellaneous registers. Address 0x10 to Address 0x1F. Set unused register to 0 if accessed. Enable VD update of miscellaneous registers, Address 0x10 to Address 0x1F. Set unused register to 0 if accessed. Enable SCK update of VDHD Registers, Address 0x20 to Address 0x22. Set unused register to 0 if accessed. Enable VD update of VDHD registers, Address 0x20 to Address 0x22. Set unused register to 0 if accessed. Enable SCK update of timing core registers, Address 0x30 to Address 0x37. Set unused register to 0 if accessed. Enable VD update of timing core registers, Address 0x30 to Address 0x37. Set unused register to 0 if accessed. Set unused register to 0 if accessed. Table 30. HPAT Registers (HPAT Registers Always Start at Address 0x800) Address 0x00 0x01 0x02 0x03 0x04 Data Bit Content [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] Default Value X X X X X X X X X X X X X X X Update SCP SCP SCP SCP SCP Name HBLKTOGO1 HBLKTOGO2 Unused HBLKTOGO3 HBLKTOGO4 Unused HBLKTOGO5 HBLKTOGO6 Unused HBLKTOGE1 HBLKTOGE2 Unused HBLKTOGE3 HBLKTOGE4 Unused Description First HBLK Toggle Position for Odd Lines, or RA0H1REPA/B/C. Second HBLK Toggle Position for Odd Lines, or RA1H1REPA/B/C. Set unused bits to 0. Third HBLK Toggle Position for Odd Lines, or RA2H1REPA/B/C. Fourth HBLK Toggle Position for Odd Lines, or RA3H1REPA/B/C. Set unused bits to 0. Fifth HBLK Toggle Position for Odd Lines, or RA4H1REPA/B/C. Sixth HBLK Toggle Position for Odd Lines, or RA5H1REPA/B/C. Set unused bits to 0. First HBLK Toggle Position for Even Lines, or RA0H2REPA/B/C. Second HBLK Toggle Position for Even Lines, or RA1H2REPA/B/C. Set unused bits to 0. Third HBLK Toggle Position for Even Lines, or RA2H2REPA/B/C. Fourth HBLK Toggle Position for Even Lines, or RA3H2REPA/B/C. Set unused bits to 0. Rev. A | Page 47 of 52 AD9974 Address 0x05 0x06 0x07 0x08 0x09 0xA 0xB 0xC 0xD 0xE 0xF Data Bit Content [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [27:13] [2:0] [5:3] [8:6] [11:9] [14:12] [17:15] [19:18] Default Value X X X X X X X X X X X X X X X [20] [27:21] [12:0] [20:13] [21] [22] [27:23] [12:0] [25:13] [27:26] [27:0] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] X X X X X X X X X X X X X X X X X X X X X X X Update SCP SCP SCP SCP SCP SCP SCP SCP SCP SCP SCP Name HBLKTOGE5 HBLKTOGE6 Unused HBLKSTARTA HBLKSTARTB Unused HBLKSTARTC Unused HBLKALT_PAT1 HBLKALT_PAT2 HBLKALT_PAT3 HBLKALT_PAT4 HBLKALT_PAT5 HBLKALT_PAT6 HBLK_MODE TESTMODE Unused HBLKLEN HBLKREP HBLKMASK_H1 HBLKMASK_H2 Unused HBLKSTART HBLKEND Unused TESTMODE CLPOB0_TOG1 CLPOB0_TOG2 Unused CLPOB1_TOG1 CLPOB1_TOG2 Unused PBLK0_TOG1 PBLK0_TOG2 Unused PBLK1_TOG1 PBLK1_TOG2 Unused Description Fifth HBLK Toggle Position for Even Lines, or RA4H2REPA/B/C. Sixth HBLK Toggle Position for Even Lines, or RA5H2REPA/B/C. Set unused bits to 0. HBLK Repeat Area Start Position A. Used during HBLK Mode 2. HBLK Repeat Area Start Position B. Used during HBLK Mode 2. Set unused bits to 0. HBLK Repeat Area Start Position C. Used during HBLK Mode 2. Set unused bits to 0. HBLK Pattern 1 Order. Used during pixel mixing mode. HBLK Pattern 2 Order. Used during pixel mixing mode. HBLK Pattern 3 Order. Used during pixel mixing mode. HBLK Pattern 4 Order. Used during pixel mixing mode. HBLK Pattern 5 Order. Used during pixel mixing mode. HBLK Pattern 6 Order. Used during pixel mixing mode. HBLK Mode Selection. 0 = normal HBLK. 1 = pixel mixing mode. 2 = special pixel mixing mode. 3 = not used. Test Operation Only. Set to 0. Set unused bits to 0. HBLK Length in HBLK Alteration Modes. Number of HBLK Repetitions in HBLK Alternation Modes. Masking Polarity for H1/H3 During HBLK. Masking Polarity for H2/H4 During HBLK. Set unused bits to 0. HBLK Start Position Used in Pixel Mixing Modes. HBLK End Position Used in Pixel Mixing Modes. Set unused bits to 0. Test Operation Only. Set to 0. CLPOB0 Toggle Position 1. CLPOB0 Toggle Position 2. Set unused bits to 0. CLPOB1 Toggle Position 1. CLPOB1 Toggle Position 2. Set unused bits to 0. PBLK0 Toggle Position 1. PBLK0 Toggle Position 2. Set unused bits to 0. PBLK1 Toggle Position 1. PBLK1 Toggle Position 2. Set unused bits to 0. Name SCP0 SCP1 Unused SCP2 SCP3 Unused Description Sequence Change Position 0. Sequence Change Position 1. Set unused bits to 0. Sequence Change Position 2. Sequence Change Position 3. Set unused bits to 0. Table 31. Field Registers Address 0x00 0x01 Data Bit Content [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] Default Value X X X X X X Update VD VD Rev. A | Page 48 of 52 AD9974 Address 0x02 0x03 0x04 0x05 0x06 0x07 0x08 0x09 0xA 0xB 0xC 0xD 0xE 0xF Data Bit Content [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [27:13] [4:0] [9:5] [14:10] [19:15] [24:20] [27:25] [4:0] [9:5] [14:10] [19:15] [27:20] [27:0] [8:0] [17:9] Default Value X X X X X X X X X X X X X X X X X X X X X [27:18] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [8:0] [17:9] X X X X X X X X X X X [27:18] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] [12:0] [25:13] [27:26] X X X X X X X X X X Update VD VD VD VD VD VD VD VD VD VD VD VD VD VD Name SCP4 SCP5 Unused SCP6 SCP7 Unused SCP8 Unused HPAT_SEL0 HPAT_SEL1 HPAT_SEL2 HPAT_SEL3 HPAT_SEL4 Unused HPAT_SEL5 HPAT_SEL6 HPAT_SEL7 HPAT_SEL8 Unused Unused CLPOB_POL CLPOB_PAT Unused CLPOBMASKSTART1 CLOBMASKEND1 Unused CLPOBMASKSTART2 CLOBMASKEND2 Unused CLPOBMASKSTART3 CLOBMASKEND3 Unused PBLK_POL PBLK_PAT Unused PBLKMASKSTART1 PBLKMASKEND1 Unused PBLKMASKSTART2 PBLKMASKEND2 Unused PBLKMASKSTART3 PBLKMASKEND3 Unused Description Sequence Change Position 4. Sequence Change Position 5. Set unused bits to 0. Sequence Change Position 6. Sequence Change Position 7. Set unused bits to 0. Sequence Change Position 8. Set unused bits to 0. Selected H-Pattern for First Region in Field. Selected H-Pattern for Second Region in Field. Selected H-Pattern for Third Region in Field. Selected H-Pattern for Fourth Region in Field. Selected H-pattern for fifth region in field. Set unused bits to 0. Selected H-Pattern for Sixth Region in Field. Selected H-Pattern for Seventh Region in Field. Selected H-Pattern for Eighth Region in Field. Selected H-Pattern for Ninth Region in Field. Set unused bits to 0. Set unused bits to 0. CLPOB Start Polarity Settings. CLPOB Pattern Selector. 0 = CLPOB0_TOG registers are used. 1 = CLPOB1_TOG registers are used. Set unused bits to 0. CLPOB Mask 1 Start Position. CLPOB Mask 1 End Position. Set unused bits to 0. CLPOB Mask 2 Start Position. CLPOB Mask 2 End Position. Set unused bits to 0. CLPOB Mask 3 Start Position. CLPOB Mask 3 End Position. Set unused bits to 0. PBLK Start Polarity Settings for Sequence 0 to Sequence 8. PBLK Pattern Selector. 0 = PBLK0_TOG registers are used. 1 = PBLK1_TOG registers are used. Set unused bits to 0 PBLK Mask Region 1 Start Position. PBLK Mask Region 1 End Position. Set unused bits to 0. PBLK Mask Region 2 Start Position. PBLK Mask Region 2 End Position. Set unused bits to 0. PBLK Mask Region 3 Start Position. PBLK Mask Region 3 End Position. Set unused bits to 0. Rev. A | Page 49 of 52 AD9974 OUTLINE DIMENSIONS A1 CORNER INDEX AREA 9.10 9.00 SQ 8.90 10 9 8 7 6 5 4 3 2 1 A B BALL A1 PAD CORNER TOP VIEW C D 7.20 BSC SQ E F G H J K 0.80 BSC DETAIL A 1.40 MAX DETAIL A 0.65 MIN 0.25 MIN SEATING PLANE COPLANARITY 0.12 COMPLIANT TO JEDEC STANDARDS MO-205-AB. 012006-0 0.55 0.50 0.45 BALL DIAMETER Figure 56. 100-Lead Chip Scale Package Ball Grid Array [CSP_BGA] (BC-100-1) Dimensions shown in millimeters ORDERING GUIDE Model AD9974BBCZ 1 AD9974BBCZRL1 1 Temperature Range −25°C to +85°C −25°C to +85°C Package Description 100-Lead Chip Scale Package Ball Grid Array [CSP_BGA] 100-Lead Chip Scale Package Ball Grid Array [CSP_BGA] Z = RoHS Compliant Part. Rev. A | Page 50 of 52 Package Option BC-100-1 BC-100-1 AD9974 NOTES Rev. A | Page 51 of 52 AD9974 NOTES ©2009 Analog Devices, Inc. All rights reserved. Trademarks and registered trademarks are the property of their respective owners. D05955-0-10/09(A) Rev. A | Page 52 of 52