! "" Data Manual May 2001 Digital Audio Video SLAS342 IMPORTANT NOTICE Texas Instruments and its subsidiaries (TI) reserve the right to make changes to their products or to discontinue any product or service without notice, and advise customers to obtain the latest version of relevant information to verify, before placing orders, that information being relied on is current and complete. All products are sold subject to the terms and conditions of sale supplied at the time of order acknowledgment, including those pertaining to warranty, patent infringement, and limitation of liability. TI warrants performance of its products to the specifications applicable at the time of sale in accordance with TI’s standard warranty. Testing and other quality control techniques are utilized to the extent TI deems necessary to support this warranty. Specific testing of all parameters of each device is not necessarily performed, except those mandated by government requirements. Customers are responsible for their applications using TI components. In order to minimize risks associated with the customer’s applications, adequate design and operating safeguards must be provided by the customer to minimize inherent or procedural hazards. TI assumes no liability for applications assistance or customer product design. TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, or other intellectual property right of TI covering or relating to any combination, machine, or process in which such products or services might be or are used. TI’s publication of information regarding any third party’s products or services does not constitute TI’s approval, license, warranty or endorsement thereof. Reproduction of information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied by all associated warranties, conditions, limitations and notices. Representation or reproduction of this information with alteration voids all warranties provided for an associated TI product or service, is an unfair and deceptive business practice, and TI is not responsible nor liable for any such use. Resale of TI’s products or services with statements different from or beyond the parameters stated by TI for that product or service voids all express and any implied warranties for the associated TI product or service, is an unfair and deceptive business practice, and TI is not responsible nor liable for any such use. Also see: Standard Terms and Conditions of Sale for Semiconductor Products. www.ti.com/sc/docs/stdterms.htm Mailing Address: Texas Instruments Post Office Box 655303 Dallas, Texas 75265 Copyright 2001, Texas Instruments Incorporated Contents Section 1 2 3 Title Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1 Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2 Functional Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 Terminal Assignments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.4 Ordering Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.5 Abbreviations Used in This Document . . . . . . . . . . . . . . . . . . . . . . . . . . 1.6 Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.7 THS8083A Terminal Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Analog Channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Clamping Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Composite Sync Slicer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3.1 Implementation When Using Channel 1 Sync Slicing From Ch1 as Selected by CS_SEL, or on Prerevision A Silicon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Programmable Gain Amplifier (PGA) . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5 A/D Converter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6 PLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6.1 Analog PLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6.2 Digital PLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7 Output Formatter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8 Power Down . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.9 Input Mode Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.10 ADC Readback Over I2C Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Register Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 I2C Protocol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.1 Write Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1.2 Read Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Register Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.1 Register Name: TERM_CNT_0 . . . . . . . . . . . . . . . . . . . . . . . 3.2.2 Register Name: TERM_CNT_1 . . . . . . . . . . . . . . . . . . . . . . 3.2.3 Register Name: NOM_INC_0 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.4 Register Name: NOM_INC_1 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.5 Register Name: NOM_INC_2 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.6 Register Name: NOM_INC_3 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.7 Register Name: NOM_INC_4 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.8 Register Name: VCODIV . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.9 Register Name: SELCLK . . . . . . . . . . . . . . . . . . . . . . . . . . . . Page 1–1 1–1 1–3 1–4 1–4 1–5 1–5 1–5 2–1 2–1 2–1 2–2 2–4 2–4 2–4 2–4 2–4 2–5 2–8 2–8 2–9 2–9 3–1 3–1 3–1 3–2 3–5 3–5 3–5 3–5 3–5 3–6 3–6 3–6 3–6 3–7 iii 4 iv 3.2.10 Register Name: PHASESEL . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.11 Register Name: PLLFILT . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.12 Register Name: HS_WIDTH . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.13 Register Name: VS_WIDTH . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.14 Register Name: SYNC_CTRL . . . . . . . . . . . . . . . . . . . . . . . . 3.2.15 Register Name: LD_THRES . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.16 Register Name: PLL_CTRL . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.17 Register Name: HS_COUNT_0 . . . . . . . . . . . . . . . . . . . . . . 3.2.18 Register Name: HS_COUNT_1 . . . . . . . . . . . . . . . . . . . . . . 3.2.19 Register Name: VS_COUNT_0 . . . . . . . . . . . . . . . . . . . . . . 3.2.20 Register Name: VS_COUNT_1 . . . . . . . . . . . . . . . . . . . . . . 3.2.21 Register Name: DTO_INC_0 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.22 Register Name: DTO_INC_1 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.23 Register Name: DTO_INC_2 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.24 Register Name: DTO_INC_3 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.25 Register Name: DTO_INC_4 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.26 Register Name: SYNC_DETECT . . . . . . . . . . . . . . . . . . . . . 3.2.27 Register Name: CLP_CTRL . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.28 Register Name: CLP_START_0 . . . . . . . . . . . . . . . . . . . . . . 3.2.29 Register Name: CLP_START_1 . . . . . . . . . . . . . . . . . . . . . . 3.2.30 Register Name: CLP_STOP_0 . . . . . . . . . . . . . . . . . . . . . . . 3.2.31 Register Name: CLP_STOP_1 . . . . . . . . . . . . . . . . . . . . . . . 3.2.32 Register Name: CH1_CLP . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.33 Register Name: CH1_COARSE . . . . . . . . . . . . . . . . . . . . . . 3.2.34 Register Name: CH1_FINE . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.35 Register Name: CH2_CLP . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.36 Register Name: CH2_COARSE . . . . . . . . . . . . . . . . . . . . . . 3.2.37 Register Name: CH2_FINE . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.38 Register Name: CH3_CLP . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.39 Register Name: CH3_COARSE . . . . . . . . . . . . . . . . . . . . . . 3.2.40 Register Name: CH3_FINE . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.41 Register Name: PIX_TRAP_0 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.42 Register Name: PIX_TRAP_1 . . . . . . . . . . . . . . . . . . . . . . . . 3.2.43 Register Name: PWDN_CTRL . . . . . . . . . . . . . . . . . . . . . . . 3.2.44 Register Name: AUX_CTRL . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.45 Register Name: CH1_RDBK . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.46 Register Name: CH2_RDBK . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.47 Register Name: CH3_RDBK . . . . . . . . . . . . . . . . . . . . . . . . . 3.2.48 Register Name: OFM_CTRL . . . . . . . . . . . . . . . . . . . . . . . . . Parameter Measurement Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Timing Diagram—24-Bit Parallel Mode . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Timing Diagram—16-Bit Parallel Mode . . . . . . . . . . . . . . . . . . . . . . . . . 4.3 Timing Diagram—48-Bit Interleaved Mode . . . . . . . . . . . . . . . . . . . . . . 4.4 Timing Diagram—48-Bit Parallel Mode . . . . . . . . . . . . . . . . . . . . . . . . . 3–7 3–7 3–7 3–8 3–8 3–8 3–9 3–9 3–10 3–10 3–10 3–10 3–10 3–11 3–11 3–11 3–11 3–11 3–12 3–12 3–12 3–12 3–13 3–13 3–13 3–13 3–13 3–13 3–14 3–14 3–14 3–14 3–14 3–14 3–15 3–15 3–16 3–16 3–16 4–1 4–1 4–2 4–3 4–4 5 Electrical Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1 5.1 Definition of Test Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1 5.2 Absolute Maximum Ratings Over Operating Free-Air Temperature Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–2 5.3 Recommended Operating Conditions Over Operating Free-Air Temperature Range, TA = 0°C to 70°C . . . . . . . . . . . . . . . . . 5–2 5.3.1 Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–2 5.3.2 Analog and Reference Inputs . . . . . . . . . . . . . . . . . . . . . . . . . 5–2 5.3.3 Digital Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–2 5.4 Electrical Characteristics Over Recommended Operating Free-Air Temperature Range, TA = 0°C to 70°C . . . . . . . . . . . . . . . . . . . . . . . . . 5–3 5.4.1 Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–3 5.4.2 Digital Logic Inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–3 5.4.3 Logic Outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–3 5.4.4 I2C Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–4 5.4.5 ADC Channel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–4 5.4.5.1 DC Accuracy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–4 5.4.5.2 Dynamic Performance . . . . . . . . . . . . . . . . . . . . 5–5 5.4.5.3 Clamp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–5 5.4.6 Coarse PGA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–5 5.4.7 Fine PGA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–5 5.4.8 Output Formatter/Timing Requirements . . . . . . . . . . . . . . . . 5–6 5.4.9 PLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–6 5.4.9.1 Open Loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–6 5.4.9.2 Closed Loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–7 5.4.10 Typical Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–7 6 Application Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–1 6.1 Designing With PowerPAD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–1 7 Mechanical Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7–1 Appendix A. PLL Formulas and Register Settings . . . . . . . . . . . . . . . . . . . . . . A–1 v List of Illustrations Figure Title Page 2–1 Analog Channel Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–1 2–2 Bottom-Level Clamping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–2 2–3 Mid-Level Clamping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–2 2–4 Using THS8083A With a Composite Sync . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–3 2–5 Analog PLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–5 2–6 Digital PLL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–7 2–7 Output Formatter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2–8 5–1 Input Test Waveform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–1 5–2 Power Consumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–7 5–3 Linearity of AGY Channel at 80 MSPS (external clock) . . . . . . . . . . . . . . . . . 5–8 List of Tables Table Title 2 3–1 I C Register Map . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3–2 Output Formatter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–1 Junction-Ambient and Junction-Case Thermal Resistances . . . . . . . . . . . . . vi Page 3–3 3–16 6–1 1 Introduction The THS8083A is a complete solution for digitizing video and graphic signals in RGB or YUV/YCbCr color spaces. The device supports pixel rates up to 80 MHz. Therefore, it can be used for PC graphics digitizing up to the VESA standard of XGA (1024 X 768) resolution at 75 Hz screen refresh rate, and in video environments for the digitizing of digital TV formats, including HDTV. The THS8083A is powered from a single 3.3-V supply and integrates a triple high-performance A/D converter with clamping functions and variable gain, independently programmable for each channel. The clamp timing window is provided by an external pulse or can be generated internally. The programmable gain amplifiers consist of coarse and fine gain control blocks. The THS8083A includes slicing circuitry on the Y or G input to support sync-on-green or sync-on-luminance extraction. The THS8083A also contains a completely digital PLL block consisting of phase-frequency detector (PFD), discrete time oscillator (DTO), and programmable divider to generate the (sampling) clock from the incoming horizontal sync (HS) signal, depending on the incoming video resolution. Any pixel rate can be generated in the 13–80 MHz range. Moreover, the output phase of the synthesized clock can be controlled with subpixel accuracy (31 uniform settings). Programmable time constants allow changing the PLL loop bandwidth by the integrated PLL loop filter. Alternatively, the user may bypass the PLL when an external pixel clock is available. Even then the DTO synthesized clock is still available externally and can therefore be used in other parts of the (graphics) system. Extensive PLL and input monitoring functions are integrated for typical functionality required in LCD/DMD monitor/projection systems (input format detection, autocalibration). All programming of the part is done via an industry-standard normal/fast I2C interface, which supports both reading and writing of register settings. The THS8083A is available in a space-saving TQFP 100-pin PowerPAD package. 1.1 Features The THS8083A supports the following features: • • Analog Channels – Three digitizing channels, each with independently controllable clamp, PGA, and ADC – Clamp: 256-step programmable RGB or YUV clamping during external or internal clamp timing window – PGA: 6-bit coarse/5-bit fine programmable gain amplifier – ADC: 8 bit 80 MSPS A/D converter – Composite sync: Integrated sync-on-green/sync-on-luminance extraction from green/luminance channel or from dedicated input – Support for dc and ac-coupled input signals PLL – Fully integrated digital PLL (including loop filter) for pixel clock generation – 13-80 MHz pixel clock generation from reference input – Adjustable PLL loop bandwidth for minimum jitter or fast acquisition/wide capture range modes – 5-bit programmable subpixel accurate positioning of sampling phase – Noise gates on HS and VS inputs to avoid false PLL updating PowerPAD is a trademark of Texas Instruments. 1–1 • • • 1–2 Output Formatter – Single and double pixel width output data bus for reduced board clock frequency and EMI – Support for 4:4:4 and 4:2:2 (ITU–R BT.601 style) output modes to reduce board traces – Dedicated DATACLK output for easy latching of output data System – Industry-standard normal/fast I2C interface with register readback capability – Support for input format detection via integrated monitoring of HS, VS, and pixel clock frequencies – Support for multidevice operation (master/slave operation for SXGA resolution) – Space-saving TQFP-100 pin package – Thermally-enhanced PowerPAD package for better heat dissipation Applications – LCD desktop monitors and LCD or DMD-based projection systems – Videoconferencing – PCTV set-top boxes, digital TV sets, and multimedia cards – Scan rate/image resolution converters – Video/graphics digitizing equipment (RGB or YUV-based) 1.2 Functional Block Diagram THS8083A-95 CS_IN CS AVSS_REF VMID VREFTO_CHn VREFBO_CHn OE Bandgap Reference Slicer VCM AVDD_REF Reference Generator Programmable Variable Gain Amplifier Clamp CH1_IN ADC CH1A (0–7) ADC1_OUT CH1B (0–7) AVDD_CH1 AVSS_CH1 Reference Generator Programmable Variable Gain Amplifier Clamp CH2_IN ADC ADC2_OUT CH2A (0–7) Output Formatter CH2B (0–7) AVDD_CH2_3 AVSS_CH2_3 Reference Generator CH3_IN Programmable Variable Gain Amplifier Clamp CH3A (0–7) ADC ADC3_OUT CH3B (0–7) EXT_CLP HS VS Noise Gates Clamp Timing Generator VSS CLP DVDD DVSS LOCK PLL PFD_FREEZE Control Interface (I2C) EXT_ADCCLK XTL1 XTL2 DVDD_PLL DHS DTOCLK3 DVSS_PLL ADCCLK2 AVDD_PLL AVSS_PLL I2CA SCAN_TEST RESET SDA SCL DATACLK1 1–3 1.3 Terminal Assignments 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 CH3A0 CH3A1 CH3A2 CH3A3 CH3A4 CH3A5 CH3A6 CH3A7 CH2A0 CH2A1 CH2A2 CH2A3 CH2A4 CH2A5 CH2A6 CH2A7 CH1A0 CH1A1 CH1A2 CH1A3 CH1A4 CH1A5 CH1A6 CH1A7 DVDD 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 HS VS SCL SDA I2CA CH3B0 CH3B1 CH3B2 CH3B3 CH3B4 CH3B5 CH3B6 CH3B7 CH2B0 CH2B1 CH2B2 CH2B3 DVDD DVSS XTL1_MCLK DVSS_PLL XTL2 DVDD_PLL AVSS_PLL AVDD_PLL 100 99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 80 79 78 77 76 LOCK EXT_CLP PFD_FREEZE VREFTO_CH3 VREFBO_CH3 CH3_IN NC NC AVDD_CH2_3 AVSS_CH2_3 VREFTO_CH2 VREFBO_CH2 CH2_IN NC NC AVDD_CH1 AVSS_CH1 VREFTO_CH1 VREFBO_CH1 CH1_IN NC CS_IN/TEST2 CS/TEST1 SCAN_TEST RESET TQFP PowerPAD PACKAGE (TOP VIEW) 1.4 Ordering Information PACKAGED DEVICES TA 1–4 TQFP–100 Maximum clock frequency 80 MSPS 0°C to 70°C THS8083APZP 75 74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 AVDD_REF AVSS_REF VMID VCM OE CH1B7 CH1B6 CH1B5 CH1B4 CH1B3 CH1B2 CH1B1 CH1B0 CH2B7 CH2B6 CH2B5 CH2B4 DVSS DVDD EXT_ADCCLK DHS DATACLK1 ADCCLK2 DTOCLK3 VSS 1.5 Abbreviations Used in This Document PGA Programmable gain amplifier PLL I2C Phase-locked loop EMI Electro-magnetic interference NTSC National Television Systems Committee PAL Phase alternating line DTV Digital TV VBI Vertical blanking interval CS Composite sync Inter-IC interface 1.6 Conventions Throughout this document, the term YUV refers to a video/graphics signal, consisting of three components, of which one component (Y) has its blanking level corresponding to the bottom level of the video signal range. The other two components (U&V) have their blanking level at the mid-scale of the video signal range, because U&V are color difference signals and thus, can go positive or negative with respect to blanking. YUV, therefore, should not be restricted to NTSC/PAL component formats, but also includes baseband component video formats used in DTV that should in a strict sense be denoted as analog YCbCr or YPbPr. The term RGB refers to a video/graphics signal, consisting of three components, of which all components have their blanking level corresponding to the bottom level of the video signal range. Therefore, it relates to both RGB PC formats as well as red-green-blue video component signals, sometimes denoted as GBR instead of RGB in video broadcast environments. 1.7 THS8083A Terminal Functions TERMINAL NAME NO. I/O/B† TYPE‡ DESCRIPTION POWER SUPPLIES AVSS_PLL 24 I A Analog ground for PLL (XTL oscillator and analog PLL) AVDD_PLL 25 I A Analog supply (3.3 V) for analog PLL DVSS_PLL 21 I A Digital ground for digital PLL DVDD_PLL 23 I A Digital supply (3.3 V) for digital PLL AVSS_CH1 84 I A Analog ground for A/D channel 1 AVDD_CH1 85 I A Analog supply (3.3 V) for A/D channel 1 AVSS_CH2_3 91 I A Analog ground for A/D channel 2 and channel 3 AVDD_CH2_3 92 I A Analog supply (3.3 V) for A/D channel 2 and channel 3 DVDD 18, 50, 57 I A Digital supply for all logic, except digital PLL DVSS 19, 58 I A Digital ground for all logic, except digital PLL VSS 51 I A Substrate ground AVDD_REF 75 I A Analog supply (3.3 V) for voltage and current reference generator A Analog ground (3.3 V) for voltage and current reference generator AVSS_REF 74 I † I = input to device: O = output from device ‡ A = analog pin: D = digital pin B = bidirectional 1–5 1.7 THS8083A Terminal Functions Order (Continued) TERMINAL NAME NO. I/O/B† TYPE‡ DESCRIPTION CLOCK I/O XTL1_MCLK 20 I A Master crystal connection 1 (connects 14.318-MHz crystal) or master clock input (at 14.318 MHz) XTL2 22 DATACLK1 54 O A O D 53 O D DTOCLK3 52 O D Master crystal connection 2 (connects 14.318-MHz crystal) 1st clock output: DATACLK1 The rising edge of this clock can be used by an external device to clock in THS8083A output data in all modes (see output timing diagrams in Section 4 for more details). 2nd clock output: ADCCLK This clock output is equal to the clock of the ADC converter, optionally inverted and/or divided-by-2. 3rd clock output: DTOCLK. This clock output is the output of the DTO. ADCCLK2 EXT_ADCCLK 56 I D External clock input for A/D channels, at pixel clock frequency CH1_IN 81 I A Analog input channel 1. Since this channel includes the composite sync slicer and is not downsampled in 4:2:2 mode, this channel should be used for green or luminance input, if any of these features are used. CH2_IN 88 I A Analog input channel 2. In YUV 4:2:2 sampling mode, Pb should be connected to this input to generate a ITU.BT-601 style output. CH3_IN 95 I A Analog input channel 3. In YUV 4:2:2 sampling mode, Pr should be connected to this input to generate a ITU.BT-601 style output. VREFBO_CH1 82 B A Reference voltage bottom output channel 1. In normal operation: output. For a specific configuration, this terminal becomes an input terminal (see Powerdown section in Functional Description). VREFTO_CH1 83 B A Reference voltage top output channel 1. In normal operation it is an output. For a specific configuration, this terminal becomes an input terminal (see Powerdown section in Functional Description). VREFBO_CH2 89 B A Reference voltage bottom output channel 2. See VREFBO_CH1. VREFTO_CH2 90 B A Reference voltage top output channel 2. See VREFTO_CH1. VREFBO_CH3 96 B A Reference voltage bottom output channel 3. See VREFBO_CH1. VREFTO_CH3 97 B A Reference voltage top output channel 3. See VREFTO_CH1. VMID 73 B A Midlevel input range (input common mode). In normal operation it is an output. For a specific configuration, this terminal becomes an input terminal (see Powerdown section in Functional Description). VCM 72 O A Common mode voltage output (approximately 1.5 V) ANALOG SIGNAL I/O DIGITAL SIGNAL I/O CH1A0 42 O D Display output channel 1, bus A, bit 0 (LSB) CH1A1 43 O D Display output channel 1, bus A, bit 1 CH1A2 44 O D Display output channel 1, bus A, bit 2 CH1A3 45 O D Display output channel 1, bus A, bit 3 CH1A4 46 O D Display output channel 1, bus A, bit 4 CH1A5 47 O D Display output channel 1, bus A, bit 5 CH1A6 48 O D Display output channel 1, bus A, bit 6 CH1A7 49 O D Display output channel 1, bus A, bit 7 (MSB) CH1B0 63 O D Display output channel 1, bus B, bit 0 (LSB) CH1B1 64 O D Display output channel 1, bus B, bit 1 † I = input to device: O = output from device ‡ A = analog pin: D = digital pin 1–6 B = bidirectional 1.7 THS8083A Terminal Functions Order (Continued) TERMINAL NAME NO. I/O/B† TYPE‡ DESCRIPTION DIGITAL SIGNAL I/O (Continued) CH1B2 65 O D Display output channel 1, bus B, bit 2 CH1B3 66 O D Display output channel 1, bus B, bit 3 CH1B4 67 O D Display output channel 1, bus B, bit 4 CH1B5 68 O D Display output channel 1, bus B, bit 5 CH1B6 69 O D Display output channel 1, bus B, bit 6 CH1B7 70 O D Display output channel 1, bus B, bit 7 (MSB) CH2A0 34 O D Display output channel 2, bus A, bit 0 (LSB) CH2A1 35 O D Display output channel 2, bus A, bit 1 CH2A2 36 O D Display output channel 2, bus A, bit 2 CH2A3 37 O D Display output channel 2, bus A, bit 3 CH2A4 38 O D Display output channel 2, bus A, bit 4 CH2A5 39 O D Display output channel 2, bus A, bit 5 CH2A6 40 O D Display output channel 2, bus A, bit 6 CH2A7 41 O D Display output channel 2, bus A, bit 7 (MSB) CH2B0 14 O D Display output channel 2, bus B, bit 0 (LSB) CH2B1 15 O D Display output channel 2, bus B, bit 1 CH2B2 16 O D Display output channel 2, bus B, bit 2 CH2B3 17 O D Display output channel 2, bus B, bit 3 CH2B4 59 O D Display output channel 2, bus B, bit 4 CH2B5 60 O D Display output channel 2, bus B, bit 5 CH2B6 61 O D Display output channel 2, bus B, bit 6 CH2B7 62 O D Display output channel 2, bus B, bit 7 (MSB) CH3A0 26 O D Display output channel 3, bus A, bit 0 (LSB) CH3A1 27 O D Display output channel 3, bus A, bit 1 CH3A2 28 O D Display output channel 3, bus A, bit 2 CH3A3 29 O D Display output channel 3, bus A, bit 3 CH3A4 30 O D Display output channel 3, bus A, bit 4 CH3A5 31 O D Display output channel 3, bus A, bit 5 CH3A6 32 O D Display output channel 3, bus A, bit 6 CH3A7 33 O D Display output channel 3, bus A, bit 7 (MSB) CH3B0 6 O D Display output channel 3, bus B, bit 0 (LSB) CH3B1 7 O D Display output channel 3, bus B, bit 1 CH3B2 8 O D Display output channel 3, bus B, bit 2 CH3B3 9 O D Display output channel 3, bus B, bit 3 CH3B4 10 O D Display output channel 3, bus B, bit 4 CH3B5 11 O D Display output channel 3, bus B, bit 5 CH3B6 12 O D Display output channel 3, bus B, bit 6 CH3B7 13 O D † I = input to device: O = output from device ‡ A = analog pin: D = digital pin Display output channel 3, bus B, bit 7 (MSB) B = bidirectional 1–7 1.7 THS8083A Terminal Functions Order (Continued) TERMINAL NAME NO. I/O/B† TYPE‡ DESCRIPTION DIGITAL CONTROL I/O SCL 3 B D Clock for I2C. Although the device is an I2C slave, this signal can be held low by the device to signal contention, therefore it is flagged bidirectional. SDA 4 B D Serial data for I2C I2CA 5 I D Address select for I2C 0 = LSB of device address 0 1 = LSB of device address 1 EXT_CLP 99 I D External clamp timing pulse. Positive polarity required. HS 1 I D Reference clock input for PLL (horizontal sync input). Polarity selectable via I2C register <HS_POL>. 5-V tolerant input VS 2 I D Vertical sync input. Polarity selectable via I2C register <VS_POL>. 5 V tolerant input DHS 55 O D Display horizontal sync. This output can be generated as either a delayed version of input HS or as output pulse from the PLL feedback divider. See Display Horizontal Sync section in Functional Description. CS/TEST1 78 O A/D Composite sync output. This output produces a 3-V logic-compatible sliced output of CH1 or CS_IN, depending on CS_SEL (see CS_IN/TEST2 terminal). When present and enabled, CS carries the embedded composite sync. See Composite Sync Slicer section in Functional Description. For TI internal testing, this pin can also be configured as a test pin. Leave unconnected when CS output signal is not used. CS_IN/TEST2 79 I A Composite sync slicing input. When selected by CS_SEL register, the signal on this pin is clamped to blanking level according to the clamp timing pulse and sliced approximately 150 mV below this clamped level to produce a composite sync output available on CS/TEST1. This pin can also be configured as a test pin for TI internal testing. Leave unconnected when CS input signal is not used. LOCK 100 O D Lock detect output 0 = unlocked 1 = locked PFD_FREEZE 98 I D Freezes the PLL output frequency by stopping the PFD output (i.e., keeping last increment to DTO). See section 2.3 Composite Sync Slicer. 0 = updating 1 = frozen OE 71 I D Output enable for data output busses A and B. Data outputs are active only when OE = L and the corresponding bus is active for the current output formatter mode (register OFM_CTRL). When data outputs are not active or when DVDD = 0 V, data output is Hi-Z. The clock outputs are not affected by OE. 0 = enabled 1 = disabled RESET 76 I D General chip reset (active low). The reset is a synchronous reset. Therefore, a master clock on XTL1–MCLK needs to be present for proper reset. TEST I/O CS/TEST1 78 O A/D See previous description for this terminal under DIGITAL CONTROL I/O. TEST2 79 O A/D See previous description for this terminal under DIGITAL CONTROL I/O. SCAN_TEST 77 I D Input for scan-path activation: 0 = disabled 1 = enabled. This pin MUST be tied low for normal operation and is of use for TI internal testing only. UNUSED PINS NC 80, 86, 87, 93, 94 I † I = input to device: O = output from device ‡ A = analog pin: D = digital pin 1–8 A Not connected. Tie to a fixed high or low level on board. B = bidirectional 2 Functional Description 2.1 Analog Channel The THS8083A contains three identical analog channels that are independently programmable. Each channel consists of a clamping circuit, a programmable gain amplifier, and an A/D converter. 2.2 Clamping Circuit The purpose of clamping is to provide the input signal with a known dc-value. Typically, video signals are ac-coupled into the part. The signal needs to be level-shifted to fall in the reference voltage range (VREFB...VREFT) of the A/D converter. By supplying a programmable clamp, the user can shift the input signal with respect to the A/D range. This has the same effect as keeping the input signal constant and applying offset to both A/D reference voltages while keeping the VREFT–VREFB difference equal. However, no external adjustments are needed with this implementation. For video, the clamping circuit can only be active during the non-active video portion of each line to avoid changes in brightness along the line. Clamping is done during the horizontal blanking interval, either on the back porch of sync or during the sync tip (in the case of a sync present on at least one of the video channels). If HS is carried on a separate line, as is typically the case for PC graphics, clamping is done during blanking. When the Y or G input channel contains an embedded sync, then alternatively clamping can be done during the sync-tip or during the front or back porch of sync. Only clamping during front- or back-porch of sync is supported on the THS8083A, since it is expected that the input signal level during clamping, of which position and width are determined by the clamp timing pulse (as shown later) corresponds to the blanking level. Since the blanking level for RGB type inputs corresponds to a low output code of the A/D, it makes sense to center the clamp range around an A/D output code of 0. The user can adjust this level up or down, symmetrically around 0. If the clamping is set such that the blanking level corresponds to a level below 0, the A/D output is clipped at code 0. Reference Level CLP PGA 2 PGA 1 VIN 8 ADC CC Bottom/Mid Clamp DAC Reference Level 8 Offset DAC Clamp Control 6 PGA Gain Control Figure 2–1. Analog Channel Architecture 2–1 In the case of YUV input signals, blanking levels for U and V correspond to the mid-level analog input. To handle these signals the clamping range should be centered on the mid-level output code of the A/D. The clamp code is 8 bits wide and spans 128 ADC output codes (a 2 LSB change to clamp code corresponds nominally to 1 LSB change in ADC output). The programmed clamp code is independent of the PGA setting (see later). This ensures independent brightness/clamping control. The clamp pulse defines the timing window during which the clamp circuit is internally enabled, and is either generated externally and supplied to the device, or it can be internally generated. In the latter case, the user can program both the position and width of the clamp pulse with respect to the horizontal sync (HS) input. CLAMP CODE CLIP 255 255 ADC Output = +63 = 0 CLAMP CODE = –64 CLIP 255 255 +63 = +63 191 ADC Output Code Range 0 = –64 +63 191 Code Range 0 63 0 CLIP 0 =0 63 –64 –64 0 CLIP 0 VIN VIN CLP PULSE Influence of changing clamp codes on A/D output, while keeping PGA gain setting constant, in bottom-level clamp mode Figure 2–2. Bottom-Level Clamping CLP PULSE Influence of changing clamp codes on A/D output, while keeping PGA gain setting constant, in midlevel clamp mode Figure 2–3. Mid-Level Clamping 2.3 Composite Sync Slicer The THS8083A includes a circuit that compares the input signal on Ch.1, or on the dedicated CS_IN input, to a level 150 mV below the blanking level. This slicer outputs a 3-V compatible digital output on the composite sync (CS) pin. The intended use of this circuit is for input video signals that have an embedded (negative or trilevel) sync. This is the case for workstation-type input signals or the DTV analog interface that mandates sync-on-Y. Since the sync amplitude is ~300 mV, the slicing level is at about 50% of the sync level. When enabled, the CS output is available even when the device is powered down. CS outputs the extracted composite sync. Since the PLL is prevented from updating its phase detector while the PFD_FREEZE pin is kept high, the user asserts PFD_FREEZE during the VBI (when CS has multiple transitions per line). This puts the PLL in free-run. While it cannot be assured with devices that have analog PLL’s, the digital PLL in the THS8083A is assured to keep a constant output frequency and avoid frequency drift while the PLL is in free-run. There is also no maximum on the time that PFD_FREEZE can be kept asserted to still keep a stable PLL output frequency. In this case, the CS output can be directly connected to the THS8083A’s HS input for purposes of locking the PLL. However, the frequency monitoring of HS, which works off signal edges, produces invalid numbers on those lines where CS is present because of the multiple low-high transitions on these lines. 2–2 Alternatively, if an external sync separator is present that generates HS and VS from CS, the separated signals can be fed to the corresponding inputs on the THS8083A and PFD_FREEZE can be left unused. As long as HS has one pulse per line, the PLL can lock correctly and the HS frequency monitoring register will return the correct value. VS is only used by the field/frame frequency monitoring register and this will return the correct value as long as VS has one pulse per field/frame. Both options are shown in Figure 2–4. Option 1: Using PFD_FREEZE Frame Period PFD_FREEZE High in VBI on Lines Where CS Has Multiple Rising/Falling Edges Per Line PFD_FREEZE Ext. Logic CS THS8083A HS Option 2: Using HS Derived From CS PFD_FREEZE CS Sync. Separator THS8083A HS VS Figure 2–4. Using THS8083A With a Composite Sync as Note that the slicer only works when no video levels are lower than the blanking level and when the internal clamp circuit is used. This is normally satisfied for G and Y channels, but not for U and V channels. To prevent unnecessary toggling of the CS output signal, the CS output is switched off (i.e. HI-Z) automatically when mid-level clamping is chosen for channel 1 (i.e., CLP1_RG=1 in register <CLP_CTRL>). The source for the slicer is either Ch1 or the dedicated input CS_IN, as selected by register CS_SEL. It is recommended to use the dedicated input by ac-coupling the Y/G signal input to both Ch1 and CS_IN (using independent coupling capacitors). The THS8083A performs independent clamping on both inputs. NOTE:In this revision-A silicon, PDF_FREEZE keeps the DTO output frequency constant and disables the phase-frequency detector (PFD) from internally updating its error value at every active edge of HS (this was not the case in prerevision-A versions). Therefore, when PFD_FREEZE is asserted, the PLL effectively ignores incoming pulses on the HS terminal. In this revision-A silicon, users do not need to provide an external dc biasing to the Y/G channel since now a dedicated input terminal for composite sync slicing is provided. This was not the case in prerevision-A versions, where the sync could only be extracted from Channel1 and a clamp diode had to be used to establish an initial clamping level. This method is described in the following section. When using the dedicated CS_IN terminal, the G/Y signal should be ac-coupled to this terminal independently from Ch1, i.e., the G/Y signal is routed via a coupling capacitor to CS_IN and via a second coupling capacitor to Ch1. This is because the THS8083A imposes an independent (programmable) clamp level for the sync input. 2–3 2.3.1 Implementation When Using Channel 1 Sync Slicing From Ch1 as Selected by CS_SEL, or on Prerevision A Silicon To support sync-on-Y/sync-on-G extraction, users should provide an external dc biasing to the Y/G channel. This can be done by establishing a dc clamp through a diode with its cathode connected to the ac-coupling capacitor (at the side of the THS8083A) on the AGY channel and its anode connected to a dc level. Since the slicing level is approximately 1.35 V and the sync amplitude is –300 mV, the negative sync-tip should be clamped by the diode to a level of approximately 1.2 V. For example, using a Schottky switching diode (type 1N5711) with a maximum low forward voltage drop of 0.4 V, the dc level at the anode can be approximately 1.6 V. This level can be derived through a resistive voltage divider off the power supply. 2.4 Programmable Gain Amplifier (PGA) Each video channel is passed through a programmable gain amplifier, to provide a full-scale signal to each A/D. The user can change this gain via register programming. A gain change becomes effective immediately. The range of the PGA is such that an input ac range from 0.4 Vpp to 1.2 Vpp can be scaled to ADC full scale, by maximum gain and minimum gain settings respectively. The PGA is split into a 6-bit coarse gain control and 5-bit fine gain control. Their combination leads to a PGA resolution of better than 1 LSB on the ADC output code. The bandwidth of the PGA is by design constant, resulting in a constant analog video input bandwidth. The coarse PGA, with its 64 settings, covers a 4/3 x to 4x gain change, used for a 0.4 V (0.4 Vpp × 4 = 1.6 Vpp) respectively 1.2 Vpp (1.2 Vpp × 4/3 = 1.6 Vpp) input range swing. While an amplifier with variable gain implements the coarse PGA, the fine PGA is implemented by slightly changing the top and bottom reference levels that are also independently controllable for each ADC channel. The fine range, with its 32 settings, covers a range of 16 LSBs. The fine and coarse PGA settings can be combined into a single PGA gain formula as follows: GAIN = (4/3 + C/24)(1 + (F–15)/512) Where C is the coarse gain setting (0..63) and F the fine gain setting (0..31). 2.5 A/D Converter The A/D converter’s switched-capacitor single-pipeline CMOS architecture combines excellent signal-to-noise characteristics with a very wide 3-dB analog input bandwidth of typically 500 MHz. The A/D block contains an internal reference voltage generator, providing stable bottom and top references derived from an internal bandgap reference. The reference voltages are made available externally. The THS8083A supports both dc and ac-coupled inputs (clamping disabled). With dc-coupling, available external references can be used to level-shift the input signal. The A/D converter is assured up to 80 MSPS with no missing codes. The sampling clock of the A/D converter is either externally fed or internally generated by the PLL. 2.6 PLL The PLL is a fully contained functional block consisting of: • An analog PLL operating at a fixed output frequency of N times the master (crystal) clock frequency • A digital PLL containing a digital phase-frequency detector (PFD), a discrete time oscillator (DTO), a digital loop filter, a feedback divider, a programmable clock output divider, and a programmable phase shifter 2.6.1 Analog PLL The analog PLL generates a high-frequency internal clock used by the DTO in the digital PLL to derive the pixel output frequency with programmable phase. The reference signal for this PLL is the master clock frequency supplied on the XTL1-MCLK terminal. 2–4 Two options exist for connecting a master clock: • A crystal can be connected between the XTL1-MCLK and XTL2 terminals. The device provides internal oscillator circuitry. • A 3.3-V CMOS/TTL clock signal can be connected to XTL1-MCLK from an external oscillator. In this case XTL2 must be left unconnected. The port is designed to operate from a master clock frequency of 14.31818 MHz, which is a standard frequency in video applications: 4x is the subcarrier frequency for NTSC. Many low-cost crystals are available for this frequency. The default internal oscillator operates at 8x the master clock frequency, or about 114 MHz. This setting of 8x, which is the value of the feedback divider in the analog PLL loop, is programmable (VCODIV register value). Normally this remains as the default 8x value. Users can change this value when a master clock of a different frequency is connected. In this case care should be taken to keep the internal high-frequency clock (i.e., master clock frequency x analog feedback divider) lower than 120 MHz. The higher this internal frequency, the better the frequency resolution of the DTO. When a crystal is used as the master clock source, it is not advised to use another frequency than the recommended 14.31818 MHz, since the internal oscillator circuitry is not production tested at other frequencies. If another master clock is used, it is recommended to drive XTL1–MCLK by a direct clock signal. VCODIV should be programmed such that the internal clock remains close to but less than 120 MHz. 14.31818 MHz PhaseFrequency Detector Loop Filter VCO VCOCLK (To Digital PLL) Programmable Divider 3 VCODIV Figure 2–5. Analog PLL 2.6.2 Digital PLL The digital PLL loop derives the ADC (pixel) clock frequency from the high-speed internal clock. A DTO generates an output frequency from a user-programmable DTO increment. To operate over the 13–80 MHz range, an extra DTO clock output divider can be switched in. Appendix A shows the formula that relates the frequency of the internal high-speed clock, the DTO increment value, and the DTO clock output divider to the PLL output frequency. The PLL output, after the clock divider, is sent to the programmable feedback divider (TERM_CNT register value). This value is typically programmed with the number of total pixels per line for a given video/graphics format. The output of this divider is then one input to the phase-frequency detector. Its other input is typically the horizontal sync (HS) reference of a graphics/video signal. HS needs to be provided as a separate TTL/CMOS type signal to the dedicated input terminal. See the Composite Sync Slicer section to use the PLL in the case of input signals with a composite sync. The polarity of HS is programmable (HS_POL register value). Both HS and VS inputs on the THS8083A can accept a 3-V and a 5-V logic-compliant signal. On the HS input, as on the VS input, a digital noise gate can be optionally switched in (HS_MS and VS_MS register values, respectively). The user can program the minimum number of clock cycles that have to be present in HS and VS before they are interpreted as a valid HS and VS. This avoids having any spikes being interpreted as an active HS and falsely updating the PLL. 2–5 The PFD produces a digital error value, signaling the phase/frequency difference between the HS input and the divided PLL output clock. The integrated digital PLL loop filter subsequently filters this error value. This filter consists of proportional and integrator (accumulator) parts. Gains of both parts are programmable (GAIN_N and GAIN_P register values), each with eight settings. The higher the programmed value, the higher the gain in either the proportional or integrator portions of the filter, which translates into a wider capture range and faster acquisition but also into higher steady-state jitter. The PFD also provides a LOCK output on a dedicated terminal. This output has programmable hysteresis (LD_THRES register value). Details are explained in the Register Description section. The LOCK output is made available on a dedicated pin so that the user could implement additional functionality before using this output (e.g., implement the sticky nature of an unlock condition by routing it through an external set/reset flip-flop). By integrating the loop filter and making it programmable, the user can trade off both at runtime depending on the quality of the incoming HS signal (inaccurate frequency, jitter content). The filtered phase/frequency error value is now used to correct the programmed nominal DTO increment (NOM_INC register value). This updated DTO increment (DTO_INC register value) is used by the DTO and determines the instantaneous DTO output frequency. By making DTO_INC available as a read-only register, the user can read out via I2C and calculate the instantaneous frequency of the DTO-generated clock. Because of the digital nature of the PLL, the loop can be opened while still keeping an accurate frequency output. Therefore, the PLL can also be used as a frequency synthesizer without any HS reference. This is done by disabling the PFD (PFD_DISABLE register value). This keeps DTO_INC always equal to NOM_INC, thereby producing a DTO output frequency always equal to the desired programmed frequency, irrespective of HS. There is a second option to operate in open loop. In some video/graphics modes, no valid HS is present during a part of the frame/field period, typically during some lines of the VBI (vertical blanking interval). In order to have an accurate PLL output clock and avoid clock drift, the PFD output needs to be held constant during this time. The PFD FREEZE pin provides this option. Asserting this pin freezes DTO_INC to its present value, thereby producing a constant PLL output clock frequency, not necessarily equal to the nominal desired frequency programmed by NOM_INC. Together with the composite sync slicer, this feature allows using the PLL with input signals containing embedded composite sync with minimal external logic. See the Composite Sync Slicer section. The phase of the PLL generated clock can be programmed in 31 uniform steps over a single clock period (360/31 = 11.6 degrees phase resolution) so that the sampling phase of the ADC can be accurately controlled. In addition to sourcing the ADC channel clock from the PLL, an external pixel clock can be used (from terminal EXT_ADCCLK). If configured this way (via SEL_ADCCLK register value), a clock signal of the required sampling frequency should be applied to EXT_ADCCLK; this signal, instead of the PLL generated clock, is routed to the ADC channels. In this case no phase control is available on the external clock signal. Still, the internal PLL can be used and its output made available externally as explained below. This means two clock domains can be implemented on the THS8083A: one is externally fed, and the other, possibly asynchronous to the first, generated by the internal PLL. This provides considerable flexibility in the design of video/graphics equipment that implements scaling and frame rate conversion. 2–6 PFD_FREEZE DTO_DIS VCOCLK (From Analog PLL) Compensated in Output Formatter for Pipeline Data Delay. Then Output on Terminal DHS With Polarity Determined by <DHS_POL>. MUX 1 HS_POL HS_MS 1 HS POL DISABLE_ PFD 1 1 DHS_MODE SELCLK DIV2 3 1 1 Noise Gate PROG. LOOP FILTER PhaseFrequency Detector 8 HS_WIDTH DTO Phase Selector DIV 3P 3 33 GAIN_N GAIN_P NOM_INC 5 PHASESEL D I V 2 PLLCLK MUX LD_THRESH 8 D I V 2 TERM_CNT LOCK INV2 to ADC 12 ADCCLK2 1 1 SEL_ADCCLK Programmable Divider Lock Detection Hysteresis I N V I N V 1 1 DIV3 INV3 ADCCLK1 (see NOTE) DTOCLK3 EXT_ADCCLK NOTE: ADCCLK1 is used by the output formatter to generate the DATACLK1 output. Figure 2–6. Digital PLL The device provides three clock outputs. One output signal, DATACLK1, is derived from the ADC clock output. It is actually equal to the sampling clock, but compensated in phase so that its rising edge always corresponds to the center valid region of the output data. Output data timing (setup/hold) is specified with respect to this rising edge. Therefore, DATACLK1 is typically used for clocking the THS8083A’s output data. The frequency of DATACLK1 is either equal to, or one half the sampling clock, depending on the operation mode of the output formatter. When the THS8083A is clocked with an external sampling clock, this external clock is used as the source to generate DATACLK1 in the output formatter. The second clock output, ADCCLK2, is equal to the ADC sampling clock, but can optionally be divided by 2 and inverted. The third clock output, DTOCLK3, is always derived from the PLL output clock, irrespective of the use of an external sampling clock on EXT_ADCCLK. So, when operating with an external sampling clock, the DTOCLK3 output can be used to generate a second, possibly asynchronous, clock signal in either open or closed loop operation locked to a reference HS input. Also, DTOCLK3 can be optionally divided by 2 and inverted. The divide and invert functions are implemented to enable a two-part master/slave operation in case sampling speeds higher than 80 MSPS are required. In this case, the master uses its PLL to generate a line-locked clock, and its inverse is used by the second slave device as an external sampling clock. 2–7 2.7 Output Formatter This block enables either a 4:4:4 24-bit output or a 4:4:4 48-bit output at half the pixel clock, or a 4:2:2 16-bit output useful for YUV digitizing (ITU.BT-601 style). In the latter case, an 8-bit port is used for the Y output while a second 8-bit port is used alternately for Cr or Cb. As per ITU BT-601, Cb is the first video data word for each line, as shown in Figure 2-7. The first color sample after an incoming HS is Cb. The output signal DHS is synchronized to the first pixel of a line and can therefore be used to uniquely identify Cb from Cr output data in down-sampled modes. X Sampling Format Cr (R-Y or V) Y Y 3 Channels Cb (or B-Y or U) Output Formatter Combines Ob and Cr On Single Output Bus Cb Cr Cb Cr Cb 2 Channels On Ch1 Bus A Output On Ch2 Bus A Output Y Y t Y Y Y Other Outputs HI-Z Output Format Figure 2–7. Output Formatter 2.8 Power Down Three power down modes are defined in the I2C power-down register, : • Chip power down: PWDN_ALL When PWDN_ALL=1, all analog circuits are powered down except the internal bandgap reference, the circuit that generates the clamping voltages and the sync reference voltage. All these are kept active for the composite sync slicer that remains active during power down. The clock frequency of the digital circuitry is lowered to reduce power consumption when in power down. • Internal reference power down: PWDN_REF When PWDN_REF=1, bottom and top references (VREFB, VREFT) on all channels become inputs and should be externally driven. • Bandgap reference power down: PWDN_BGAP When PWDN_BGAP=1, the internal bandgap reference voltage is inactive and terminal VMID should be externally driven. Additionally, the DTO circuitry can be disabled: • 2–8 DTO power down: DTO_DIS When DTO_DIS=1, the DTO frequency is lowered to reduce power dissipation. This feature can be activated when an external sampling clock is used (EXT_ADCCLK). 2.9 Input Mode Detection The THS8083A supports detection of the graphics input format in cooperation with an external microcontroller. Via the microcontroller interface, the period of incoming HS and VS signals (HS_COUNT, VS _COUNT register values), the frequency of the DTO clock (DTO_INC register value), and the PLL lock condition (terminal LOCK) can be measured. 2.10 ADC Readback Over I2C Interface The ADC output data on each of the three channels can be sampled at a programmable position on each line (PIXTRAP register value) and latched into pixel readback registers (CH< n >_RDBK register values) that can be read by the microcontroller at lower speed via the I2C interface. Programming to read back during the horizontal blanking interval can be a test for accurate positioning of the blanking level. 2–9 2–10 3 Register Definition 3.1 I2C Protocol The THS8083A is a slave I2C device that supports both write and read. As shown in Table 3–1, I 2C Register Map, there are some status control registers that can only be read. The device can support FAST I2C mode (SCL up to 400 kHz) when the DTO clock is running at over 25 MHz; at lower DTO frequencies, only NORMAL I2C mode (SCL up to 100 kHz) is supported. To discriminate between write and read operations, the device is addressed at separate device addresses. There is an automatic internal subaddress increment counter to efficiently write/read multiple bytes in the register map during one write/read operation. Furthermore, bit 1 of the I2C device address is dependent upon the I2CA pin setting, as follows: If address selecting pin I2CA = 0, then Write address is 40 hex (01000000) Read address is 41 hex (01000001) If address selecting pin I2CA = 1, then Write address is 42 hex (01000010) Read address is 43 hex (01000011) 3.1.1 S Write Format Slave address(w) A Subaddress A Data0 A …… S Start condition Slave address(w) 0100000 (0x40) if I2CA=0 / 01000010 (0x42) if I2CA=1 A Acknowledge. It is generated by THS8083A. Subaddress Subaddress of first register to write, length: 1 byte Data0 First byte of data Data(N–1) Nth byte of data P Stop condition Data(N–1) A P 3–1 3.1.2 Read Format First write the subaddress, where data needs to be read out, to the THS8082/3A in the following format: S Slave address(w) A Slave address(r) A Subaddress A P Then: S DataN AM Data(N+1) AM …… NAM P S Start condition Slave address(r) 01000001 (0x41) if I2CA=0 / 01000011 (0x43) if I2CA=1 A Acknowledge, generated by the THS8082/3A. If transmission is successful, then A = 0, else A = 1. AM Acknowledge, generated by a master NAM Not acknowledge, generated by a master Subaddress Subaddress of the first register to read, length = one byte Data0 First byte of the data read Data(N–1) Nth byte of the data read P Stop condition In both write and read operations, the subaddress is incremented automatically when multiple bytes are written/read. So, only the first subaddress needs to be supplied to the THS8082/3A. R/W registers can be written and read. R registers are read-only. 3–2 Table 3–1. I2C Register Map REGISTER NAME R/W SUB ADDRESS Bit 7 Bit 6 Bit 5 Bit 4 TERM_CNT_0 R/W 00 TERM_CNT7 TERM_CNT6 TERM_CNT5 TERM_CNT4 TERM_CNT_1 R/W 01 NOM_INC_0 R/W 02 NOM_INC7 NOM_INC6 NOM_INC5 NOM_INC4 NOM_INC_1 R/W 03 NOM_INC15 NOM_INC14 NOM_INC13 NOM_INC12 NOM_INC_2 R/W 04 NOM_INC23 NOM_INC22 NOM_INC21 NOM_INC20 NOM_INC_3 R/W 05 NOM_INC31 NOM_INC30 NOM_INC29 NOM_INC28 NOM_INC_4 R/W 06 VCODIV R/W 07 SELCLK R/W 08 PHASESEL R/W 09 PLLFILT R/W 0A GAIN_N2 GAIN_N1 GAIN_N0 GAIN_P2 GAIN_P1 GAIN_P0 HS_WIDTH R/W 0B HS_WIDTH7 HS_WIDTH6 HS_WIDTH5 HS_WIDTH4 HS_WIDTH3 HS_WIDTH2 HS_WIDTH1 HS_WIDTH0 VS_WIDTH R/W 0C VS_WIDTH7 VS_WIDTH6 VS_WIDTH5 VS_WIDTH4 VS_WIDTH3 VS_WIDTH2 VS_WIDTH1 VS_WIDTH0 SYNC_CTRL R/W 0D HS_POL HS_MS VS_POL VS_MS LD_THRES R/W 0E LD_THRES3 LD_THRES2 LD_THRES1 LD_THRES0 PLL_CTRL Bit 3 Bit 2 Bit 1 Bit 0 TERM_CNT3 TERM_CNT2 TERM_CNT1 TERM_CNT0 TERM_CNT11 TERM_CNT10 TERM_CNT9 TERM_CNT8 NOM_INC3 NOM_INC2 NOM_INC1 NOM_INC0 NOM_INC11 NOM_INC10 NOM_INC9 NOM_INC8 NOM_INC19 NOM_INC18 NOM_INC17 NOM_INC16 NOM_INC27 NOM_INC26 NOM_INC25 NOM_INC24 NOM_INC32 VCODIV2 PHASE_SEL4 LD_THRES7 LD_THRES6 HS_COUNT7 HS_COUNT6 VS_COUNT7 VS_COUNT6 PHASE_SEL3 PHASE_SEL2 VCODIV1 VCODIV0 SELCLK1 SELCLK0 PHASE_SEL1 PHASE_SEL0 LD_THRES5 LD_THRES4 DISABLE_PFD SEL_ADCCLK INV2 DIV2 INV3 DIV3 HS_COUNT5 HS_COUNT4 HS_COUNT3 HS_COUNT2 HS_COUNT1 HS_COUNT0 HS_COUNT11 HS_COUNT10 HS_COUNT9 HS_COUNT8 VS_COUNT5 VS_COUNT4 VS_COUNT3 VS_COUNT2 VS_COUNT1 VS_COUNT0 VS_COUNT11 VS_COUNT10 VS_COUNT9 VS_COUNT8 R/W 0F HS_COUNT_0 R 10 HS_COUNT_1 R 11 VS_COUNT_0 R 12 VS_COUNT_1 R 13 DTO_INC_0 R 14 DTO_INC7 DTO_INC6 DTO_INC5 DTO_INC4 DTO_INC3 DTO_INC2 DTO_INC1 DTO_INC0 DTO_INC_1 R 15 DTO_INC15 DTO_INC14 DTO_INC13 DTO_INC12 DTO_INC11 DTO_INC10 DTO_INC9 DTO_INC8 DTO_INC_2 R 16 DTO_INC23 DTO_INC22 DTO_INC21 DTO_INC20 DTO_INC19 DTO_INC18 DTO_INC17 DTO_INC16 DTO_INC_3 R 17 DTO_INC31 DTO_INC30 DTO_INC29 DTO_INC28 DTO_INC27 DTO_INC26 DTO_INC25 DTO_INC24 DTO_INC_4 R 18 DTO_INC32 SYNC_DETECT R 19 NO_SYNC Reserved 1A-1F NOTE: Blank register bits in this table are ignored upon write. When read they return 0. 3–3 3–4 Table 3–1. I2C Register Map (continued) REGISTER NAME R/W SUB ADDRESS CLP_CTRL R/W 20 CLP_START_0 R/W 21 CLP_START_1 R/W 22 CLP_STOP_0 R/W 23 CLP_STOP_1 R/W 24 CH1_CLP R/W 25 CH1_COARSE R/W 26 CH1_FINE R/W 27 CH2_CLP R/W 28 CH2_COARSE R/W 29 CH2_FINE R/W 2A CH3_CLP R/W 2B CH3_COARSE R/W 2C CH3_FINE R/W 2D PIX_TRAP_0 R/W 2E PIX_TRAP_1 R/W 2F PWDN_CTRL R/W 30 AUX_CTRL R/W 31 CH1_RDBK R 32 CH1_RDBK7 CH2_RDBK R 33 CH2_RDBK7 CH3_RDBK R 34 CH3_RDBK7 Reserved OFM_CTRL Bit 7 Bit 6 CLP_SEL CLP1_EN CLP1_RG CLP2_EN CLP2_RG CLP3_EN CLP3_RG CLP_START7 CLP_START6 CLP_START5 CLP_START4 CLP_START3 CLP_START2 CLP_START1 CLP_START0 CLP_START11 CLP_START10 CLP_START9 CLP_START8 CLP_STOP7 CLP_STOP6 CLP_STOP5 CLP_STOP4 CLP_STOP3 CLP_STOP2 CLP_STOP1 CLP_STOP0 CLP_STOP11 CLP_STOP10 CLP_STOP9 CLP_STOP8 CH1_CLP7 CH2_CLP7 CH3_CLP7 PIX_TRAP7 CH1_CLP6 CH2_CLP6 CH3_CLP6 PIX_TRAP6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 CH1_CLP5 CH1_CLP4 CH1_CLP3 CH1_CLP2 CH1_CLP1 CH1_CLP0 CH1_COARSE5 CH1_COARSE4 CH1_COARSE3 CH1_COARSE2 CH1_COARSE1 CH1_COARSE0 CH1_FINE0 CH1_FINE4 CH1_FINE3 CH1_FINE2 CH1_FINE1 CH2_CLP5 CH2_CLP4 CH2_CLP3 CH2_CLP2 CH2_CLP1 CH2_CLP0 CH2_COARSE5 CH2_COARSE4 CH2_COARSE3 CH2_COARSE2 CH2_COARSE1 CH2_COARSE0 CH2_FINE0 CH2_FINE4 CH2_FINE3 CH2_FINE2 CH2_FINE1 CH3_CLP5 CH3_CLP4 CH3_CLP3 CH3_CLP2 CH3_CLP1 CH3_CLP0 CH3_COARSE5 CH3_COARSE4 CH3_COARSE3 CH3_COARSE2 CH3_COARSE1 CH3_COARSE0 CH3_FINE4 CH3_FINE3 CH3_FINE2 CH3_FINE1 CH3_FINE0 PIX_TRAP4 PIX_TRAP3 PIX_TRAP2 PIX_TRAP1 PIX_TRAP0 PIX_TRAP11 PIX_TRAP10 PIX_TRAP9 PIX_TRAP8 PWDN_REF PWDN_BGAP DTO_DIS PIX_TRAP5 PWDN_ALL CS_SEL CS_DIS TEST2 TEST1 TEST0 TACT CH1_RDBK6 CH1_RDBK5 CH1_RDBK4 CH1_RDBK3 CH1_RDBK2 CH1_RDBK1 CH1_RDBK0 CH2_RDBK6 CH2_RDBK5 CH2_RDBK4 CH2_RDBK3 CH2_RDBK2 CH2_RDBK1 CH2_RDBK0 CH3_RDBK6 CH3_RDBK5 CH3_RDBK4 CH3_RDBK3 CH3_RDBK2 CH3_RDBK1 CH3_RDBK0 DHS_MODE DHS_POL OFM_MODE1 OFM_MODE0 35-3F R/W 40 NOTE: Blank register bits in this table are ignored upon write. When read they return 0. 3.2 Register Description Register values after reset/at power up/after power down mode: The default value with each register shows the start-up condition after general chip reset. The register state after power up is undefined i.e., the device requires a reset after power up (RESET low) to put all registers in their default states. The value of these registers is preserved in all power-down modes (i.e. after power down the register values are identical as when entering power down); they do not return to their default values under this condition. In order for the device to reset correctly, a master clock signal needs to be applied during reset from either a clock signal on XTL1–MCLK or a crystal connected between XTL1–MCLK and XTL2. The reset signal needs to be at least 5 clock cycles wide. Default values: The default values for this device are set for [email protected] MHz. 3.2.1 Register Name: TERM_CNT_0 Subaddress: 00 (R/W) MSB LSB TERM_CNT7 TERM_CNT6 TERM_CNT5 TERM_CNT4 TERM_CNT3 TERM_CNT2 TERM_CNT1 TERM_CNT0 TERM_CNT[7..0]: TERM_CNT[11..0] sets the number of pixels per line. Controls the digital PLL feedback divider. Default: 0x20 3.2.2 Register Name: TERM_CNT_1 Subaddress: 01 (R/W) MSB LSB X X X X TERM_CNT11 TERM_CNT10 TERM_CNT9 TERM_CNT8 TERM_CNT[11..8]: See register TERM_CNT_0. Default: 0x5 Default TERM_CNT: 0x520 = 1312 pixels/line (XGA@75 Hz) 3.2.3 Register Name: NOM_INC_0 Subaddress: 02 (R/W) MSB LSB NOM_INC7 NOM_INC6 NOM_INC5 NOM_INC4 NOM_INC3 NOM_INC2 NOM_INC1 NOM_INC0 NOM_INC[7..0]: NOM_INC[32..27]: integer part of DTO increment value NOM_INC[26..0] : fractional part of DTO increment value (See Appendix A for how to calculate the increment) Default: 0x16 3.2.4 Register Name: NOM_INC_1 Subaddress: 03 (R/W) MSB LSB NOM_INC15 NOM_INC14 NOM_INC13 NOM_INC12 NOM_INC11 NOM_INC10 NOM_INC9 NOM_INC8 NOM_INC[15..8]: See register NOM_INC_0. Default: 0x8A 3–5 3.2.5 Register Name: NOM_INC_2 Subaddress: 04 (R/W) MSB LSB NOM_INC23 NOM_INC22 NOM_INC21 NOM_INC20 NOM_INC19 NOM_INC18 NOM_INC17 NOM_INC16 NOM_INC[23..16]: See register NOM_INC_0. Default: 0x2E 3.2.6 Register Name: NOM_INC_3 Subaddress: 05 (R/W) MSB LSB NOM_INC31 NOM_INC30 NOM_INC29 NOM_INC28 NOM_INC27 NOM_INC26 NOM_INC25 NOM_INC24 NOM_INC[31..24]: See register NOM_INC_0. Default: 0xBA 3.2.7 Register Name: NOM_INC_4 Subaddress: 06 (R/W) MSB LSB X X X X X X X NOM_INC32 NOM_INC32: See register NOM_INC_0 Default: 0x00 NOTE: The default value for NOM_INC is 0xBA2E8A16. As shown in Appendix A, this can be calculated to correspond to a DTO output frequency of 78.75 MHz (XGA@75Hz). IMPORTANT: To properly update the increment, it is required to successively program NOM_INC_0 to NOM_INC_4 and then repeat the programming of the two last bytes NOM_INC3 and NOM_INC4 in this order. This properly sets the DTO to the new frequency. 3.2.8 Register Name: VCODIV Subaddress: 07 (R/W) MSB LSB X X X X X VCODIV2 VCODIV1 VCODIV0 VCODIV[2..0]: Divider in analog PLL loop. Determines the internal master clock frequency as VCODIV x master clock frequency (from XTL1–MCLK/XTL2). Default: 0x03, corresponds to an analog multiplier of 8, producing an internal nominal frequency of 8x14.31818 MHz 3–6 VCO_DIV[2..0] ANALOG PLL MULTIPLIER 000 5 001 6 010 7 011 (default) 8 100 9 101 10 110 11 111 12 3.2.9 Register Name: SELCLK Subaddress: 08 (R/W) MSB LSB X X X X X X SELCLK1 SELCLK0 SELCLK[1..0]: Selects a clock divider on the DTO output, as shown below: Default: 0x01, corresponds to DTO divider = 2 Depending on the desired output clock frequency, SELCLK must be programmed as follows for VCODIV=8: SEL_CLK[1..0] DIVIDER CLKDIV Output Clock Range (MHz) 00 Illegal – 01 (default) 2 107 – 57.5 10 4 57.5 – 28.5 11 8 28.5 – 14.25 When an output frequency lower than 14.25 MHz is desired, VCODIV needs to be lowered to 7. For a given VCODIV and desired output frequency, the NOM_INC setting changes as follows according to the formula in Appendix A. 3.2.10 Register Name: PHASESEL Subaddress: 09 (R/W) MSB LSB X X X PHASESEL4 PHASESEL3 PHASESEL2 PHASESEL1 PHASESEL0 PHASESEL[4..0]: Sets the phase for the DTO clock output Default: 0x10, corresponding to a phase shift = 180 degrees 3.2.11 Register Name: PLLFILT Subaddress: 0A (R/W) MSB LSB X X GAIN_N2 GAIN_N1 GAIN_N0 GAIN_P2 GAIN_P1 GAIN_P0 GAIN_N[2..0]: PLL gain control: Sets the loop filter proportional time constant Default: 0x7 (highest gain – lowest time constant) GAIN_P[2..0]: PLL gain control: Sets the loop filter integrator time constant Default: 0x7 (highest gain – lowest time constant) NOTE: The higher the PLL gain setting, the less critical the initial DTO programming becomes since the device will have a wider lock-in range. However, once lock is acquired, any jitter on HS is amplified. Therefore, for high jitter sources, it is recommended to apply more filtering once lock is acquired to filter out this HS jitter. 3.2.12 Register Name: HS_WIDTH Subaddress: 0B (R/W) MSB LSB HS_WIDTH7 HS_WIDTH6 HS_WIDTH5 HS_WIDTH4 HS_WIDTH3 HS_WIDTH2 HS_WIDTH1 HS_WIDTH0 HS_WIDTH[7..0]: Sets the width in pixels for HS detection. If the width of the incoming HS is less than this number, it is ignored. The width in pixels of an incoming HS is incremented at each pixel following the active edge (of which the polarity can be programmed, see HS_POL) Default: 0x00 3–7 3.2.13 Register Name: VS_WIDTH Subaddress: 0C (R/W) MSB LSB VS_WIDTH7 VS_WIDTH6 VS_WIDTH5 VS_WIDTH4 VS_WIDTH3 VS_WIDTH2 VS_WIDTH1 VS_WIDTH0 VS_WIDTH[7..0]: Sets the width in lines for VS detection. If the width of the incoming VS is less than this number, it is ignored. The width in lines of an incoming VS is incremented each time a valid HS is detected during a VS active. Therefore, the programmed width is the minimum number of horizontal syncs spanned by the active duration of VS. Default: 0x00 3.2.14 Register Name: SYNC_CTRL Subaddress: 0D (R/W) MSB LSB X X X X HS_POL HS_MS VS_POL VS_MS HS_POL: Controls the polarity of the incoming HS 0 = positive polarity (default) 1 = negative polarity HS_MS: Controls the mux selection for activating the noise filter on incoming HS 0 = noise filter disabled (default) 1 = noise filter enabled VS_POL: Controls the polarity of the incoming VS 0 = positive polarity (default) 1 = negative polarity VS_MS: Controls the mux selection for activating the noise filter on incoming VS 0 = noise filter disabled (default) 1 = noise filter enabled 3.2.15 Register Name: LD_THRES Subaddress: 0E (R/W) MSB LSB LD_THRES7 LD_THRES6 LD_THRES5 LD_THRES4 LD_THRES3 LD_THRES2 LD_THRES1 LD_THRES0 LD_THRES[7..0]: Sets hysteresis for PLL lock-detection output. An internal counter counts the number of subsequent lines onto which lock is found, as follows: for each line (HS) on which the PFD finds that the PLL is locked, the counter is incremented by 1. The counter clips at 255 maximum. For each line (HS) that the PLL is not locked to, the counter is decremented by 8. This counter starts from 0. Lock is signaled externally (via the LOCK_DETECT output) when this internal counter holds a value higher than <LD_THRESHOLD>. Unlock is signaled externally when this internal counter holds a value less than or equal to <LD_THRESHOLD>. So, a value of 255 will never assert the lock signal, although the PLL might be locked internally. NOTE: the higher this value is set, the more critical the PFD will be to signal lock. Therefore, this value must be lower for high jitter HS inputs than for high quality sources. Default: 0x10 = 16 3–8 3.2.16 Register Name: PLL_CTRL Subaddress: 0F (R/W) MSB LSB X X DISABLE_PFD SEL_ADCCLK INV2 DIV2 INV3 DIV3 DISABLE_PFD: Disables updating of the DTO increment (i.e., keeps DTO output frequency constant and independent of the incoming HS frequency). This effect is similar as opening the PLL loop. 0 = PFD enabled 1 = PFD disabled (default): the DTO runs at a constant frequency, as determined by NOM_INC. This means the output frequency returns to the nominal value and further updating of the DTO output frequency is avoided (the PLL loop is open). This is chosen as the default mode to avoid false random frequency changes by the DTO caused by noise on the HS input. In normal operation the microprocessor periodically checks the SYNC_DETECT register. If sync is present/absent, then the PFD is enabled/disabled so, frequency drift is avoided when no input signal is present. Still the panel can be driven then by data with a nominal pixel frequency. SEL_ADCCLK: Selects the PLL clock or the clock signal on the EXT_ADCCLK pin, as the clock source for the ADC channels 0: internal clock selected (default) 1: external clock selected INV2 : Selects inverting or noninverting clock output on ADCCLK2 output pin 0: the output is not inverted (default) with respect to the internal ADCCLK1 clock 1: the output is inverted with respect to the internal ADCCLK1 clock DIV2: Enables divide-by-2 function on the clock output of ADCCLK2 0: divide by 2 mechanism is disabled (default) 1: divide by 2 mechanism is enabled INV3: Selects inverting or noninverting output on DTOCLK3, with respect to the internal DTOCLK3 clock 0: the output is not inverted (default) 1: the output is inverted DIV3: Enables divide-by-2 function on the clock output of DTOCLK3 0: divided by 2 mechanism is disabled (default) 1: divided by 2 mechanism is enabled 3.2.17 Register Name: HS_COUNT_0 Subaddress: 10 (R) MSB LSB HS_COUNT7 HS_COUNT6 HS_COUNT5 HS_COUNT4 HS_COUNT3 HS_COUNT2 HS_COUNT1 HS_COUNT0 HS_COUNT[7..0] HS_COUNT[11..0] holds the last horizontal sync period number (i.e., the number of pixel clock cycles between the last two HS occurrences). The device updates the value at each active edge of HS. Internal arbitration logic avoids potential read errors between the register contents and the asynchronous I2C bus. This value can be read by the microcontroller to derive the line frequency of the incoming video/graphics format. Default: (changed during operation) 3–9 3.2.18 Register Name: HS_COUNT_1 Subaddress: 11 (R) MSB LSB X X X X HS_COUNT11 HS_COUNT10 HS_COUNT9 HS_COUNT8 HS_COUNT[11..8]: See register HS_COUNT_0 Default: (changed during operation) 3.2.19 Register Name: VS_COUNT_0 Subaddress: 12 (R) MSB LSB VS_COUNT7 VS_COUNT6 VS_COUNT5 VS_COUNT4 VS_COUNT3 VS_COUNT2 VS_COUNT1 VS_COUNT0 VS_COUNT[7..0]: VS_COUNT[11..0] holds the last vertical sync period number (i.e., the number of line periods between the last two VS occurrences). The device updates the value at each active edge of VS. Internal arbitration logic avoids potential read errors between the register contents and the asynchronous I2C bus. This value can be read by the microcontroller to derive the frame rate of the incoming video/graphics format. Default: (changed during operation) 3.2.20 Register Name: VS_COUNT_1 Subaddress: 13 (R) MSB LSB X X X X VS_COUNT11 VS_COUNT10 VS_COUNT9 VS_COUNT8 VS_COUNT[11..8] See register VS_COUNT0 Default: (changed during operation) 3.2.21 Register Name: DTO_INC_0 Subaddress: 14 (R) MSB LSB DTO_INC7 DTO_INC6 DTO_INC5 DTO_INC4 DTO_INC3 DTO_INC2 DTO_INC1 DTO_INC0 DTO_INC[7..0] DTO_INC[32..0] stores the current value of the DTO increment. This can be read by the microcontroller to derive the actual pixel clock frequency. Default: (changed during operation) 3.2.22 Register Name: DTO_INC_1 Subaddress: 15 (R) MSB LSB DTO_INC15 DTO_INC14 DTO_INC13 DTO_INC12 DTO_INC[15..8]: See register DTO_INC_0 Default: (changed during operation) 3–10 DTO_INC11 DTO_INC10 DTO_INC9 DTO_INC8 3.2.23 Register Name: DTO_INC_2 Subaddress: 16 (R) MSB LSB DTO_INC23 DTO_INC22 DTO_INC21 DTO_INC20 DTO_INC19 DTO_INC18 DTO_INC17 DTO_INC16 DTOINC[23..16]: See register DTO_INC_0 Default: (changed during operation) 3.2.24 Register Name: DTO_INC_3 Subaddress: 17 (R) MSB LSB DTO_INC31 DTO_INC30 DTO_INC29 DTO_INC28 DTO_INC27 DTO_INC26 DTO_INC25 DTO_INC24 DTO_INC[31..24]: See register DTO_INC_0. Default: (changed during operation) 3.2.25 Register Name: DTO_INC_4 Subaddress: 18 (R) MSB LSB X X X X X X X DTO_INC32 DTO_INC32: See register DTO_INC_0. Default: (changed during operation) 3.2.26 Register Name: SYNC_DETECT Subaddress: 19 (R) MSB LSB X X X X X X X NO_SYNC NO_SYNC: Sync detection on HS. 0 = HS present 1 = HS missing Default: (changed during operation) 3.2.27 Register Name: CLP_CTRL Subaddress: 20 (R/W) MSB LSB CLPSEL CLPSEL CLP1_EN CLP1_RG CLP2_EN CLP2_RG CLP3_EN CLP3_RG Selects the clamp timing signal 0: internal clamp timing pulse is selected (default) 1: external clamp timing pulse is selected CLP1_EN: Enables/disables clamping on channel 1 1: enable (default) 0: disable CLP1_RG: Sets the clamp range for channel 1 1: middle range 0: bottom range (default) 3–11 CLP2_EN: Enables/disables clamping on channel 2 1: enable (default) 0: disable CLP2_RG: Sets the clamp range for channel 2 1: middle range 0: bottom range (default) CLP3_EN: Enables/disables clamping on channel 3 1: enable (default) 0: disable CLP3_RG: Sets the clamp range for channel 3 1: middle range 0: bottom range (default) 3.2.28 Register Name: CLP_START_0 Subaddress: 21 (R/W) MSB LSB CLP_START7 CLP_START6 CLP_START5 CLP_START4 CLP_START3 CLP_START2 CLP_START1 CLP_START0 CLP_START[7..0]: CLP_START[11..0] sets the pixel count value that defines the start of the internal clamping pulse. If external clamping is selected (via CLPSEL) this value has no meaning. Default: 0x2 3.2.29 Register Name: CLP_START_1 Subaddress: 22 (R/W) MSB LSB X X X X CLP_START11 CLP_START10 CLP_START9 CLP_START8 CLP_START[11..8]: See register CLP_START_0 Default: 0x00 3.2.30 Register Name: CLP_STOP_0 Subaddress: 23 (R/W) MSB LSB CLP_STOP7 CLP_STOP6 CLP_STOP5 CLP_STOP4 CLP_STOP3 CLP_STOP2 CLP_STOP1 CLP_STOP0 CLP_STOP[7..0]: CLP_STOP[11..0] sets the pixel count value that defines the end of the internal clamping pulse. If external clamping is selected (via CLPSEL) this value has no meaning. Default: 0x40 = 64 3.2.31 Register Name: CLP_STOP_1 Subaddress: 24 (R/W) MSB LSB X X X X CLP_STOP11 CLP_STOP10 CLP_STOP9 CLP_STOP8 CLP_STOP[11..8]: See register CLP_STOP_0 Default: 0x00 NOTE: A setting of about 62 clamp clock cycles is sufficient to assure enough clamp timing (>500 ns) at worst case (=highest clock frequency). 3–12 3.2.32 Register Name: CH1_CLP Subaddress: 25 (R/W) MSB LSB CH1_CLP7 CH1_CLP6 CH1_CLP5 CH1_CLP4 CH1_CLP3 CH1_CLP2 CH1_CLP1 CH1_CLP0 CH1_CLP[7..0] Programmable clamp value for channel 1 Default: 0x80 = 128 (mid-range) 3.2.33 Register Name: CH1_COARSE Subaddress: 26 (R/W) MSB LSB X X CH1_COARSE5 CH1_COARSE4 CH1_COARSE3 CH1_COARSE2 CH1_COARSE1 CH1_COARSE0 CH1_COARSE[5..0] Coarse PGA value for channel 1 Default: 0x20 = 32 (mid-range) 3.2.34 Register Name: CH1_FINE Subaddress: 27 (R/W) MSB LSB X X X CH1_FINE4 CH1_FINE3 CH1_FINE2 CH1_FINE1 CH1_FINE0 CH1_FINE[4..0] Fine PGA value for channel 1 Default: 0x10 = 16 (mid-range) 3.2.35 Register Name: CH2_CLP Subaddress: 28 (R/W) MSB LSB CH2_CLP7 CH2_CLP6 CH2_CLP5 CH2_CLP4 CH2_CLP3 CH2_CLP2 CH2_CLP1 CH2_CLP0 CH2_CLP[7..0] Programmable clamp value for channel 2 Default: 0x80 = 128 (mid-range) 3.2.36 Register Name: CH2_COARSE Subaddress: 29 (R/W) MSB LSB X X CH2_COARSE5 CH2_COARSE4 CH2_COARSE3 CH2_COARSE2 CH2_COARSE1 CH2_COARSE0 CH2_COARSE[5..0] Coarse PGA value for channel 2 Default: 0x20 = 32 (mid-range) 3.2.37 Register Name: CH2_FINE Subaddress: 2A (R/W) MSB LSB X X X CH2_FINE4 CH2_FINE3 CH2_FINE2 CH2_FINE1 CH2_FINE0 CH2_FINE[4..0] Fine PGA value for channel 2 Default: 0x10 = 16 (mid-range) 3–13 3.2.38 Register Name: CH3_CLP Subaddress: 2B (R/W) MSB LSB CH3_CLP7 CH3_CLP6 CH3_CLP5 CH3_CLP4 CH3_CLP3 CH3_CLP2 CH3_CLP1 CH3_CLP0 CH3_CLP[7..0] Programmable clamp value for channel 3 Default: 0x80 = 128 (mid-range) 3.2.39 Register Name: CH3_COARSE Subaddress: 2C (R/W) MSB LSB X X CH3_COARSE5 CH3_COARSE4 CH3_COARSE3 CH3_COARSE2 CH3_COARSE1 CH3_COARSE0 CH3_COARSE[5..0] Coarse PGA value for channel 3 Default: 0x20 = 32 (mid-range) 3.2.40 Register Name: CH3_FINE Subaddress: 2D (R/W) MSB LSB X X X CH3_FINE4 CH3_FINE3 CH3_FINE2 CH3_FINE1 CH3_FINE0 CH3_FINE[4..0] Fine PGA value for channel 3 Default: 0x10 = 16 (mid-range) 3.2.41 Register Name: PIX_TRAP_0 Subaddress: 2E (R/W) MSB LSB PIX_TRAP7 PIX_TRAP6 PIX_TRAP5 PIX_TRAP4 PIX_TRAP3 PIX_TRAP2 PIX_TRAP1 PIX_TRAP0 PIX_TRAP[7..0] PIX_TRAP[10..0] sets the pixel count value in a line to be sampled. Each <PIX_TRAP>th value on each line is stored in the CH<n>_RDBK registers Default: 0x04 3.2.42 Register Name: PIX_TRAP_1 Subaddress: 2F (R/W) MSB LSB X X X X PIX_TRAP11 PIX_TRAP10 PIX_TRAP9 PIX_TRAP8 PIX_TRAP[11..8]: See register PIX_TRAP_0 Default: 0x00 3.2.43 Register Name: PWDN_CTRL Subaddress: 30 (R/W) MSB LSB X X X PWDN_ALL X PWDN_REF PWDN_BGAP DTO_DIS PWDN_ALL Powers down complete chip excluding I2C, clamping, and composite sync slicer. Enables green mode for monitor standby. 0 = active (default) 1 = powered down 3–14 PWDN_REF Powers down internal top and bottom references for all channels (VREFT / VREFB). If powered down, enables user to supply external VREFT / VREFB references on corresponding pins. 0 = active (default) 1 = powered down PWDN_BGAP Powers down bandgap reference. If powered down, enables user to supply external VMID (input common mode voltage) on corresponding pin. 0 = active (default) 1 = powered down DTO_DIS Disables the DTO. Can be disabled when an external clock (EXT_ADCCLK) is used and the user does not intend to use the PLL output on DTOCLK3. When the PLL is active, it can be used as the clock source for the ADC channels or the ADC’s can still run from EXT_ADCCLK depending on the SEL_ADCCLK register setting. Note that when the DTO is enabled and the device is configured to use an external clock, the DTO clock is still available on the DTOCLK3 pin so it can be used as a general-purpose clock synthesizer for other parts in the system, possibly the display clock if this is different from the input pixel clock. Since the DTO is also used for internal clock generation, power should always be supplied to the PLL supply pins, even when the ADC sampling clock is fed from EXT_ADCCLK and DTO_DIS is active. 0 = active (default) 1 = powered down 3.2.44 Register Name: AUX_CTRL Subaddress: 31 (R/W) MSB LSB X X CS_SEL CS_DIS TEST2 TEST1 TEST0 TACT CS_SEL Composite sync select. Selects the source of the composite sync for slicing. 0 = from Ch1 input 1 = from CS_IN input CS_DIS Enables/disables the composite sync output on terminal CS/TEST1. The state of the CS output is also dependent on the clamp range (see Composite Sync Slicer section). 0 = enabled (default) 1 = disabled TEST[2..0] TACT Used for TI factory testing only; should not be changed from its all-0 default value. 3.2.45 Register Name: CH1_RDBK Subaddress: 32 (R) MSB LSB CH1_RDBK7 CH1_RDBK6 CH1_RDBK5 CH1_RDBK4 CH1_RDBK3 CH1_RDBK2 CH1_RDBK1 CH1_RDBK0 CH1_RDBK[7..0]: Readback register of ADC channel 1 Default: (changed during operation) 3–15 3.2.46 Register Name: CH2_RDBK Subaddress: 33 (R) MSB LSB CH2_RDBK7 CH2_RDBK6 CH2_RDBK5 CH2_RDBK4 CH2_RDBK3 CH2_RDBK2 CH2_RDBK1 CH2_RDBK0 CH2_RDBK[7..0]: Readback register of ADC channel 2 Default: (changed during operation) 3.2.47 Register Name: CH3_RDBK Subaddress: 34 (R) MSB LSB CH3_RDBK7 CH3_RDBK6 CH3_RDBK5 CH3_RDBK4 CH3_RDBK3 CH3_RDBK2 CH3_RDBK1 CH3_RDBK0 CH3_RDBK[7..0]: Readback register of ADC channel 3 Default: (changed during operation) 3.2.48 Register Name: OFM_CTRL Subaddress: 40 (R/W) MSB LSB X X X X DHS_MODE DHS_POL OFM_MODE1 OFM_MODE0 DHS_MODE Controls how DHS (display horizontal sync output) is generated. DHS can be a version of the signal on the HS input terminal, synchronized to the sampling clock and compensated for the data pipeline delay through the part (see timing diagrams). This preserves the HS width but has the disadvantage that, for some phase settings, there is a one-pixel uncertainty on the exact timing of DHS (if HS falls within setup/hold time of the input register that is clocked by the ADC sampling clock). Therefore, a second option exist to generate DHS as the output pulse of the PLL feedback divider. Since this pulse is generated once for every <TERM_CNT> cycles of the DTO clock, the uncertainty is resolved. This can avoid possible horizontal line jitter on the display system. The width of the DHS pulse is in this case always 1 ADC clock cycle, independent of the width of the incoming HS. This method also assures the generation of a DHS pulse on every line, even when no incoming HS is present or when it is filtered out by sync processing (e.g., from composite sync extraction). 0 = DHS is generated from the output of the PLL feedback divider (default) 1 = DHS is generated as a latched and delayed version of HS input DHS_POL Controls polarity of the DHS output 0 = positive polarity (default) 1 = negative polarity OFM_MODE[1..0]: Defines mode of output formatter and frequency on DATACLK1 as in Table 3–2. Table 3–2. Output Formatter OFM_MODE [1..0] DESCRIPTION DATACLK1 OUTPUT FREQUENCY 00 (default) 24-bit parallel mode: 24-bit output on bus A, bus B is Hi-Z Fs 01 16-bit mode 16-bit output on ch1 and ch2 of bus A, with data from ch2 and ch3 downsampled by 2 (parallel 4:2:2 CCIR–601 mode), bus B is Hi-Z Fs 10 48-bit interleaved mode 48-bit output on buses A and B at half sampling rate. Data on bus B shifted by 1 Fs clock. Fs/2 11 48-bit parallel mode 48-bit output on buses A and B at half sampling rate Fs/2 3–16 4 Parameter Measurement Information All timing diagrams are shown for operation with internal PLL clock at phase 0, and ADCCLK2 non-inverted and non-divided-by-2. 4.1 Timing Diagram—24-Bit Parallel Mode This mode outputs data on the three channels simultaneously in single-pixel mode. DATACLK1 is at the sampling clock frequency; output bus B remains high-impedance. ADCCLK2 pix 01 pix 02 7 ADCCLK2 Cycles Latency tsu(OUT) th(OUT) DATACLK CH1_OUTA[7..0] Last Samples From Previous Line 01 02 Last Samples From Previous Line 01 02 Last Samples From Previous Line 01 02 CH1_OUTB[7..0] CH2_OUTA[7..0] CH2_OUTB[7..0] CH3_OUTA[7..0] tPLH(OE) tPHL(OE) CH3_OUTB[7..0] 7 ADCCLK2 Cycles Latency HS DHS (DHS_POL = 0 Assumed-Inverted Polarity Otherwise) OE <DHS_MODE> = 1 –> Width Equal to Width of HS Input <DHS_MODE> = 0 –> DHS Width is 1 ADCCLK2 Period tsu(DHS) th(DHS) 4–1 4.2 Timing Diagram—16-Bit Parallel Mode This is the ITU–R.BT–601 style mode typically used in YUV operation of the part with a Y analog input connected to the Ch1 input of the THS8083A, and with Cb and Cr connected to the Ch2 and Ch3 inputs, respectively. The DATACLK1 output is at the sampling clock frequency and Ch3 remains unused. The output bus B of all channels is high impedance. The HS_D signal can be used to uniquely identify output data Cb from Cr. ADCCLK2 pix 01 Cb: From Ch.2 Input Cr: From Ch.3 Input pix 02 tsu(OUT) th(OUT) 7 ADCCLK2 Cycles Latency DATACLK CH1_OUTA[7..0] Last Samples From Previous Line 01 Last Samples From Previous Line 01(Cb) 02 03 CH1_OUTB[7..0] CH2_OUTA[7..0] CH2_OUTB[7..0] tPLH(OE) 01(Cr) 03(Cb) tPHL(OE) CH3_OUTA[7..0] CH3_OUTB[7..0] 7 ADCCLK2 Cycles Latency HS DHS (DHS_POL = 0 Assumed-Inverted Polarity Otherwise) <DHS_MODE> = 1 –> Width Equal to Width of HS Input <DHS_MODE> = 0 –> DHS Width is 1 ADCCLK2 Period tsu(DHS) th(DHS) OE 4–2 4.3 Timing Diagram—48-Bit Interleaved Mode This mode allows a double-pixel width output interface with one sampling clock period time offset between buses A and B. The DATACLK1 output is at half of the sampling clock frequency. Data on output bus A precedes data on output bus B. ADCCLK2 pix 01 pix 02 7 ADCCLK2 Cycles Latency DATACLK tsu(OUT) tsu(OUT) th(OUT) CH1_OUTA[7..0] Last Samples From Previous Line CH1_OUTB[7..0] Last Samples From Previous Line CH2_OUTA[7..0] Last Samples From Previous Line CH2_OUTB[7..0] Last Samples From Previous Line CH3_OUTA[7..0] Last Samples From Previous Line CH3_OUTB[7..0] Last Samples From Previous Line tPLH(OE) tPHL(OE) 03 02 01 04 03 02 04 03 01 04 02 7 ADCCLK2 Cycles Latency HS DHS (DHS_POL = 0 Assumed-Inverted Polarity Otherwise) 01 th(OUT) <DHS_MODE> = 1 –> Width Equal to Width of HS Input tsu(DHS) th(DHS) <DHS_MODE> = 0 –> DHS Width is 1 ADCCLK2 Period OE 4–3 4.4 Timing Diagram—48-Bit Parallel Mode This mode allows a double-pixel width output interface with no time offset between buses A and B. The DATACLK1 output is at half of the sampling clock frequency. Data on output bus A precedes data on output bus B. ADCCLK1 pix 01 pix 02 tsu(OUT) th(OUT) DATACLK CH1_OUTA[7..0] Last Samples From Previous Line 01 Last Samples From Previous Line CH1_OUTB[7..0] Last Samples From Previous Line CH2_OUTA[7..0] 01 Last Samples From Previous Line CH2_OUTB[7..0] Last Samples From Previous Line CH3_OUTA[7..0] 01 Last Samples From Previous Line CH3_OUTB[7..0] tdlh(OE) 03 05 02 04 03 05 02 04 03 05 02 04 tdhl(OE) 7 ADCCLK2 Cycles Latency HS 7 ADCCLK2 Cycles Latency DHS (DHS_POL = 0 Assumed-Inverted Polarity Otherwise) OE 4–4 <DHS_MODE> = 1 –> Width Equal to Width of HS Input tsu(DHS) th(DHS) <DHS_MODE> = 0 –> DHS Width is 1 ADCCLK2 Period 5 Electrical Specifications Electrical specifications over recommended operating conditions with Fs = 80 MSPS (unless otherwise noted) 5.1 Definition of Test Conditions 800 mVPP 3.6 µs 1/60 kHz = 16.6 µs Figure 5–1. Input Test Waveform Test condition SYSTEM_INTREF refers to: • All supplies at 3.3 V • XTL1_MCLK & XTL2 connected at 14.31818 MHz • No power downs enabled • XGA at 75-Hz operation mode, internal clock, clamping enabled, internal clamp timing, coarse and fine PGAs at midscale, bottom-level clamping, clamp code at midscale, 24-bit output mode • Identical ac-coupled 0.8 Vpp ramp-shape input on all 3 channels at 60.0-kHz line rate, as shown in Figure 5–1 • Use of internal bandgap and voltage references Test condition PLL refers to: • SYSTEM_INTREF, with an input signal other than the ramp-shape input test waveform of Figure 5–1. Test condition ADC_INTREF refers to: • All supplies at 3.3 V • Use of internal bandgap and voltage references • Use of external ADCCLK clock (SEL_ADCCLK = 1), driven at 81.92 MHz • No power downs enabled • Identical ac-coupled, 0.8 Vpp ramp-shape input on all three channels at 60.0-kHz line rate, as shown in Figure 5–1 Test condition ADC_EXTREF refers to: • ADC_INTREF, except: PWDN_BGAP = PWDN_REF = 1, VMID and VREFTO/BO driven from an external source at nominal levels Test condition ADC_PWDN refers to: • ADC_INTREF, except: PWDN_ALL = 1 5–1 5.2 Absolute Maximum Ratings Over Operating Free-Air Temperature Range (unless otherwise noted) Supply voltage range: Analog supplies (see Note 1) to AGND, Digital supplies (see Note 2) to DGND . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.5 to 4.5 V Analog supplies to digital supplies, AGND to DGND . . . . . . . . . . . . . . . –0.5 to 0.5 V Digital input voltage range to DGND, VI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.5 to DVDD + 0.5 V Analog input voltage range to AGND, VI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.5 to AVDD + 0.5 V Bandgap reference to AGND (see Note 3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.5 to AVDD + 0.5 V Reference voltage (VREFTO_CHx,VREFBO_CHx) input range to AGND, Vref (see Note 4) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –0.5 to AVDD + 0.5 V Operating free-air temperature range, TA: THS8083APZP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0°C to 70°C Storage temperature range, Tstg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . –55°C to 150°C † Stresses beyond those listed under “absolute maximum ratings” may cause permanent damage to the device. These are stress ratings only, and functional operation of the device at these or any other conditions beyond those indicated under “recommended operating conditions” is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. NOTES: 1. AVDD_PLL, AVDD_REF, AVDD_CH1, AVDD_CH2_3 2. DVDD_PLL, DVDD 3. Only input in case PWDN_BGAP=1 4. Only input in case PWDN_REF=1 5.3 Recommended Operating Conditions Over Operating Free-Air Temperature Range, TA = 0°C to 70°C (unless otherwise noted) 5.3.1 Power Supply PARAMETER MIN NOM MAX 3.0 3.3 3.6 Supply voltage, all supplies 5.3.2 UNIT V Analog and Reference Inputs (see Note 5) MIN NOM MAX UNIT Reference input voltage (top), VI(REFT) PARAMETER 1.88 1.9 1.92 V Reference input voltage (bottom), VI(REFB) 1.08 1.10 1.12 V VI(REFT) 1.2 V Analog input voltage (dc-coupled), VI(AIN) VI(REFB) Analog input voltage range, VI V NOTE 5: VREFTO_CHx and VREFBO_CHx can be inputs only when PWDN_REF=1. 5.3.3 Digital Inputs PARAMETER High-level input voltage, VIH Low-level input voltage, VIL MIN NOM MAX 2.0 DVDD DGND 0.2 x DVDD UNIT V V Clock period, tc 12.5 ns Pulse duration, clock high, tw(CLKH) 5.25 ns Pulse duration, clock low, tw(CLKL) 5.25 ns 5–2 5.4 Electrical Characteristics Over Recommended Operating Free-Air Temperature Range, TA = 0°C to 70°C (unless otherwise noted) NOTE: In order to reach stated performance levels, the device’s PowerPad feature should be thermally and electrically connected to the pcb ground plane, as described in section 6.1 Designing With PowerPad. 5.4.1 Power Supply (3.3 V All Supplies) PARAMETER TEST CONDITIONS MIN TYP MAX UNIT Analog supply ((=AVDD_CH1+AVDD_CH2_3+AVDD_PLL+AVDD_REF) AVDD CH1 AVDD CH2 3 AVDD PLL AVDD REF) ADC INTREF ADC_INTREF 325 365 mA Digital supply (=DVDD+DVDD_PLL) ( DVDD DVDD PLL) ADC INTREF ADC_INTREF 119 135 mA Total power dissipation normal operation ADC INTREF ADC_INTREF 1 47 1.47 1 65 1.65 W Total power dissipation dissipation, power down all modes ADC PWDN ADC_PWDN 385 mW 5.4.2 Digital Logic Inputs (HS, VS, SCL, SDA, I2CA, XTL1_MCLK, EXT_ADCCLK, OE, RESET, EXT_CLP) PARAMETER IIH IIL High-level input current IIL(CLK) IIH(CLK) Low-level input current, CLK (see Note 6) CI Input capacitance Low-level input current TEST CONDITIONS DVDD = 3.6 V, Digital inputs and CLK at 0 V for IIL; Digital inputs in uts and CLK at 3.6 V for IIH High-level input current , CLK (see Note 6) MIN TYP MAX UNIT –10 10 µA –10 10 µA –14 17 µA 17 µA –14 5 pF NOTE 6: Applies when XTL1_MCLK is driven by the clock signal directly. 5.4.3 Logic Outputs (SDA, CHn_OUTA[7..0], CHn_OUTB[7..0], DTOCLK3, ADCCLK2, DATACLK1, DHS, LOCK) PARAMETER TEST CONDITIONS VOH High-level output voltage DVDD = 3 V at IOH = 50 µA, Digital output forced high VOL Low-level output voltage DVDD = 3.6 V at IOL = 50 µA, Digital output forced low CO Output capacitance IOZ(H)/IOZ(L)† High-impedance-state output current MIN TYP MAX 2.9 V 0.15 5 DVDD = 3.6 V Worst case for VO = 3.6 V and VO = 0 V –10 UNIT V pF 10 µA † Tested for CHn-A[7..0] and CHn_B[7..0] only 5–3 5.4.4 I2C Interface PARAMETER VIL Low-level input voltage VIH High-level input voltage f(SCL) t(LOW) SCL clock frequency t(HIGH) th(DATA) High period of SCL Low period of SCL TEST CONDITIONS MIN TYP MAX UNIT 0.99 V 2.31 V 400†/100‡ 0 Valid for I2C fast mode su support ort only. See footnotes to SCL clock frequency. Data hold time kHz 1.3 µs 0.6 0§ µs µs tsu(DATA) Data setup time µs C(b) Capacitive load for each bus line# 400 pF † For DTO clock frequencies 25 MHz minimum (I2C fast mode) ‡ For DTO clock frequencies below 25 MHz (I2C normal mode) § The device must internally provide a hold time of 300 ns for the SDA signal (referred to VIH(min) of the SCL signal) in order to bridge the undefined region of the falling edge of SCL. ¶ If the device is used in a standard mode I2C system, the requirement of tsu(DATA)>=250 ns must be met. # Cb= total capacitance of one bus line in pF 100¶ 5.4.5 ADC Channel 5.4.5.1 DC Accuracy† MIN TYP MAX UNIT Integral nonlinearity (INL) PARAMETER PLL (see Note 7) TEST CONDITIONS 2 –2 ±1 2 LSB Differential nonlinearity (DNL) PLL (see Note 8) 1 –1 15 1.5 LSB No missing codes Gain error Assured ADC_INTREF (see Note 9) 60 mV Offset error ADC_INTREF (see Note 10) 60 mV † Assured at nominal voltage supply levels only. NOTES: 7. Integral nonlinearity (INL) —Integral nonlinearity refers to the deviation of each individual code from a line drawn from zero to full scale. The point used as zero occurs 1/2 LSB before the first code transition. The full-scale point is defined as a level 1/2 LSB beyond the last code transition. The deviation is measured from the center of each particular code to the true straight line between these two end points. 8. Differential nonlinearity (DNL)—An ideal ADC exhibits code transitions that are exactly 1 LSB apart. DNL is the deviation from this ideal value. Therefore, this measure indicates how uniform the transfer function step sizes are. The ideal step size is defined here as the step size for the device under test (i.e., last transition level – first transition level)/(2n – 2). Using this definition for DNL separates the effects of gain and offset error. A DNL of less than ±1 LSB ensures no missing codes. A DNL of less than ±1/2 LSB assures monotonic behavior. 9. Gain error—The first code transition should occur for an analog value 1/2 LSB above nominal negative full scale (the voltage applied to the REFBI terminal). The last transition should occur for an analog value 1/2 LSB below nominal positive full scale (the voltage applied to the REFTI terminal). Gain error is defined here as the deviation from the ideal location of the highest transition level on the ADC transfer function. 10. Offset error—The first code transition should occur at a level 1/2 LSB above zero. Offset is defined as the deviation of the actual first code transition from that point. 5–4 5.4.5.2 Dynamic Performance† TEST CONDITIONS PARAMETER ADC_INTREF Effective number of bits, ENOB (from SNR) Signal-to-total ratio without distortion, SNR Total harmonic distortion, THD Spurious free dynamic range, SFDR MIN TYP MAX UNIT fI = 20 MHz fI = 20 MHz 6.7 Bits 42 dB fI = 1 MHz fI = 1 MHz –49 dB 52 dB Analog input full-power bandwidth, BW (see Note 11) 500 MHz † Based on analog input voltage of 1 dB FS referenced to the full-scale input range and a clock signal with 50% duty cycle NOTE 11: Analog input bandwidth—The analog input bandwidth is defined as the maximum frequency of the input sine that can be applied to the device for which a 3 dB attenuation is observed in the reconstructed signal. 5.4.5.3 Clamp TEST CONDITIONS PARAMETER ADC_INTREF Clamp code adjustment range See Note 12 MIN TYP 100 Within 10% of final value MAX UNIT 138 LSB 1 ms Within 1 LSB of final value 2.4 ms Within 1 LSB 500 1000 05 0.5 16 1.6 Clamp acquisition time at input dc level change Input In ut level changed by 100 mV Clamp acquisition time, clamp code change Clamp changed from minimum to maximum Clamp droop error Droop between 2 clamps at 15 kHz line rate ns LSB NOTE 12: Clamp code adjustment range—A dc-input signal is applied to the device. The clamp code is changed from minimum to maximum setting. The corresponding change in the ADC output code is defined as the clamp code adjustment range. 5.4.6 Coarse PGA PARAMETER TEST CONDITIONS MIN Full-scale adjustment range 0.4 Accuracy ±6 5.4.7 TYP MAX 1.2 UNIT V LSB Fine PGA PARAMETER Full-scale adjustment range TEST CONDITIONS MIN –4 TYP MAX UNIT 8 LSB 5–5 5.4.8 Output Formatter/Timing Requirements PARAMETER TEST CONDITIONS fclk fclk Maximum conversion rate tsu(OUT) th(OUT), th(DHS) tsu(DHS) Setup time tPLH(OE) tPHL(OE) Propagation (delay) time, low-to-high MIN THS8083A TYP MAX 80 MHz Minimum conversion rate 13 With respect to 50% level of rising edge on DATACLK Hold time Setup time ns 1 ns ns 9.5 See Note 13 8.5 DATACLK1 output duty cycle 40% HS and data pipeline delay MHz 2 2 Propagation (delay) time, high-to-low-level output UNIT See Note 14 ns 60% See timing diagrams NOTES: 13. Output timing—OE timing tPLH(OE) is measured from the VIH(MIN) level of OE to the high-impedance state of the output data. The digital output load is not higher than 10 pF. OE timing tPHL(OE) is measured from the VIL(MAX) level of OE to the instant when the output data reaches VOH(min) or VOL(max)output levels. The digital output load is not higher than 10 pF. 14. Pipeline delay (latency)—The number of clock cycles between conversion initiation on an input sample and the corresponding output data being made available. Once the data pipeline is full, new valid output data are provided every clock cycle. 5.4.9 PLL 5.4.9.1 Open Loop PARAMETER DTO frequency range, f(DTO) TEST CONDITIONS THS8083A See Note 15 Instantaneous jitter, t(INS) Short-term jitter, t(JOS) Phase Increment MIN TYP 13 MAX UNIT 80 MHz 260 See Note 15 800 (p–p) 240 (rms) 11.25 Monotonic ps 1250 (p–p) 485 (rms) ps deg NOTE 15: PLL characterization: 5–6 • Instantaneous jitter is the pk-pk position variation of the clock rising edge between succeeding periods. • Short term jitter in open loop or closed loop is defined as the variation of the clock rising edge within one PLL update period (within the same video line). This can be visually measured by capturing the clock and displaying it on a digital scope with a persistency of one video line. Numerically, the time instants of the rising edges, at a defined voltage level, of a number N of clock cycles (N = 800) are captured at a high sampling rate. The average clock time period is calculated from these time instants. The deviation between each actual time instant and the ideal, based on the average clock time period, is defined as a statistically distributed jitter value along one line. This jitter is measured on both DATACLK1 and DTOCLK3 outputs. 5.4.9.2 Closed Loop PARAMETER TEST CONDITIONS f(HS) t(acq) HS locking range t(JCS) Short-term jitter t(JCL) Long-term jitter MIN TYP MAX UNIT 100 kHz 15 Lock-in time 5 See Note 16 See Note 16 ms 925 (p–p) 245 (rms) 1400 (p–p) 450 (rms) ps 1000 (p–p) 250 (rms) 1500 (p–p) 485 (rms) ps NOTE 16: PLL characterization: • Short term jitter in open loop or closed loop is defined as the variation within one PLL update period (within the same video line) of the clock rising edge. This is measured visually by capturing the clock and displaying it on a digital scope with a persistency of one video line. Numerically, the time instants of the rising edges, at a defined voltage level, of a number of clock cycles (N = 800) are captured at a high sampling rate. The average clock time period is calculated from these time instants. The deviation between each actual time instant and the ideal, based on the average clock time period, is defined as a statistically distributed jitter value along one line. This jitter is measured on both DATACLK1 and DTOCLK3 outputs. • Long term jitter in closed loop is defined as the variation over one video frame of the Nth clock rising edge on each line. This is measured by capturing the time instant at a defined level on the rising edge of the Nth clock after HS is reached on each line. The calculation uses the same principle used with short term jitter, but now takes one sample on every line and N = 800 lines. 5.4.10 Typical Plots (25°C and Measured for Standard VESA Graphics Formats) NOTE: The THS8083 is configured for each video mode with I2C register settings as specified in application note Using THS8083 for PC Graphics and Component Video Digitizing. POWER vs FREQUENCY CURRENT vs FREQUENCY 350 1600 Total Analog VCC = 3.3 V 300 1500 I – Current – mA P – Power – mW 250 1400 1300 1200 AVDD_CH1+AVDD_CH2_3 200 150 Total Digital DVDD_PLL AVDD_REF DVDD 100 1100 50 AVDD_PLL 1000 0 0 10 20 30 40 50 60 70 80 90 100 0 10 20 f – Frequency – MHz 30 40 50 60 70 80 90 100 f – Frequency – MHz Figure 5–2. Power Consumption 5–7 DNL – Differential Nonlinearity – LSB DIFFERENTIAL NONLINEARITY vs OUTPUT CODE 1.0 FS = 80 MSPS 0.8 0.6 0.4 0.2 –0.0 –0.2 –0.4 –0.6 1 13 25 37 49 61 73 85 97 109 121 133 145 157 169 181 193 205 217 229 241 253 INL – Integral Nonlinearity – LSB Output Code INTEGRAL NONLINEARITY vs OUTPUT CODE 1.5 FS = 80 MSPS 1.0 0.5 0.0 –0.5 –1.0 –1.5 1 13 25 37 49 61 73 85 97 109 121 133 145 157 169 181 193 205 217 229 241 253 Output Code Figure 5–3. Linearity of AGY Channel at 80 MSPS (external clock) 5–8 6 Application Information 6.1 Designing With PowerPAD The THS8083A is housed in a high-performance, thermally enhanced, 100-pin PowerPAD package (TI package designator: 100PZP). Use of the PowerPAD package does not require any special considerations except to note that the PowerPAD, which is an exposed die pad on the bottom of the device, is a metallic thermal and electrical conductor. Therefore, if not implementing the PowerPAD PCB features, the use of solder masks (or other assembly techniques) may be required to prevent any inadvertent shorting by the exposed PowerPAD of connection etches or vias under the package. The recommended option, however, is not to run any etches or signal vias under the device, but to have only a grounded thermal land as explained below. Although the actual size of the exposed die pad may vary, the minimum size required for the keepout area of the 100-pin PZP PowerPAD package is 5 mm × 5 mm. It is recommended that there be a thermal land, which is an area of solder-tinned-copper, underneath the PowerPAD package. The thermal land varies in size, depending on the PowerPAD package being used, the PCB construction, and the amount of heat that needs to be removed. In addition, the thermal land may or may not contain numerous thermal vias, depending on PCB construction. More information on this package and other requirements for using thermal lands and thermal vias are detailed in the TI application note PowerPAD Thermally Enhanced Package Application Report, TI literature number SLMA002, available via the TI Web pages beginning at URL: http://www.ti.com For the THS8083A, this thermal land should be grounded to the low impedance ground plane of the device. This improves thermal performance and the electrical grounding of the device. It is also recommended that the device ground terminal landing pads be connected directly to the grounded thermal land. The land size should be as large as possible without shorting device signal terminals. The thermal land may be soldered to the exposed PowerPAD using standard reflow soldering techniques. While the thermal land may be electrically floated and configured to remove heat to an external heat sink, it is recommended that the thermal land be connected to the low impedance ground plane of the device. Table 6-1 lists a comparison between the thermal resistances of the PowerPAD package (100PZP) used for this device and a regular 100-pin TQFP package. Table 6–1. Junction-Ambient and Junction-Case Thermal Resistances PowerPAD vs 100 PZP PowerPAD 100 PIN REGULAR TQFP θJA (°C/W) 100 PZP θJC (°C/W) 100 PZP θJA (°C/W) 100 pin regular θJC (°C/W) 100 pin regular AIRFLOW IN lfm 0 150 250 500 17.3 11.8 10.4 9.0 0.12 49 3 PowerPAD is a trademark of Texas Instruments. 6–1 6–2 7 Mechanical Data PZP (S-PQFP-G100) PowerPAD PLASTIC QUAD FLATPACK 0,27 0,17 0,50 75 0,08 M 51 50 76 Thermal Pad (see Note D) 26 100 0,13 NOM 25 1 12,00 TYP Gage Plane 14,20 SQ 13,80 16,20 SQ 15,80 1,05 0,95 0,25 0,15 0,05 0°–ā7° 0,75 0,45 Seating Plane 1,20 MAX 0,08 4146929/A 04/99 NOTES: A. B. C. D. All linear dimensions are in millimeters. This drawing is subject to change without notice. Body dimensions do not include mold flash or protrusion. The package thermal performance may be enhanced by bonding the thermal pad to an external thermal plane. This pad is electrically and thermally connected to the backside of the die and possibly selected leads. E. Falls within JEDEC MS-026 PowerPAD is a trademark of Texas Instruments. 7–1 7–2 Appendix A PLL Formulas and Register Settings If: F(XTL) = frequency of external crystal or master clock connected to XTL1 input of THS8083A F(VCO) = frequency of THS8083A–internal VCO F(DTO) = frequency of THS8083A–internal DTO F(DTOCLK) = frequency of externally available DTO clock output F(HS) = frequency of HS input CLKDIV = clock output divider setting VCODIV = feedback divider in THS8083A–internal analog PLL loop TERMCNT = feedback divider in THS8083A–internal digital PLL loop DTO_INC = DTO increment (when NOM_INC is programmed, DTO_INC is initialized to NOM_INC) Then: F(VCO) = F(XTL) x VCODIV F(DTO) = 32 x F(VCO) / DTO_INC F(DTOCLK) = F(DTO) / CLKDIV And, if PLL is locked: F(DTOCLK) = TERMCNT * F(HS) Summarizing: DTO_INC = [32xF(XTL)xVCODIV] / [F(DTOCLK)xCLKDIV] The formats of DTO_INC and NOM_INC: Both are 33-bit values consisting of a 6-bit integer and a 27-bit fractional part. So, in hexadecimal notation, the value is between 00.0000000hex and 3F.7FFFFFFhex. The decimal value of the increment is: <integer part>.<fractional part interpreted as integer value>x2^(–27). This means, to arrive at the decimal value of the increment: 1. Interpret the 6 MSBs as an integer value 2. Interpret the 27 LSBs expressed as a decimal integer value and multiply this by 2^(–27) to arrive at a fractional value 3. Add 1 and 2. Additional restrictions: – NOM_INC should be within the range [16,32]. – CLKDIV should be chosen for different output clock frequency ranges as shown in register map description for SEL_CLK. Examples: 1. For generating the XGA@75Hz pixel clock of 78.75 MHz, with F(XTL) = 14.381818 MHz and VCODIV=8: NOM_INC = [32x14.31818x8]/[78.75x2] = 23.272724... Since 23d = 17h and INT(0.272724… x 2^27) = 36604438d = 22E8A16 hex, this can be written as 0x17.22E8A16 or as a contiguous 33 bit number: 010111010001011101000101000010110 = 0BA2E8A16 hex, i.e., the default setting for NOM_INC (see register map). A–1 To achieve lock with an incoming HS, TERMCNT is programmed with 1312 (i.e., the total number of pixels per line in this mode). Below are the settings for popular video and graphics modes. VESA MODES Mode #1: 640x350 at 85 Hz → fpix = 31.5 MHz, N = 832 REGISTER NAME VALUE (DEC) VALUE (HEX) 832 340 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 40 03 NOM_INC_0 9c NOM_INC_1 2c NOM_INC_2 29.090905 1d.0ba2c9c ba NOM_INC_3 e8 NOM_INC_4 0 VCO_DIV 8 8 3 SEL_CLK 4 4 2 Mode #2: 640x400 at 85 Hz → fpix = 31.5 MHz, N = 832 (same settings as 640x350 at 85 Hz) Mode #3: 720x400 at 85 Hz → fpix = 35.5 MHz, N = 936 REGISTER NAME VALUE (DEC) VALUE (HEX) 936 3a8 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) a8 03 NOM_INC_0 fa NOM_INC_1 23 NOM_INC_2 25.813057 19.68123fa 81 NOM_INC_3 ce NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #4: 640x480 at 59.94 Hz → fpix = 25.175 MHz, N = 800 REGISTER NAME VALUE (DEC) VALUE (HEX) 800 320 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 20 03 NOM_INC_0 20 NOM_INC_1 56 NOM_INC_2 18.1998713 12.1995620 99 NOM_INC_3 91 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 8 8 03 Mode #5: 640x480 at 72 Hz → fpix = 31.5 MHz, N = 832 (same settings as 640x350 @ 85 Hz) A–2 Mode #6: 640x480 at 75 Hz → fpix = 31.5 MHz, N = 840 REGISTER NAME VALUE (DEC) VALUE (HEX) 840 348 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 48 03 NOM_INC_0 9c NOM_INC_1 2c NOM_INC_2 29.090905 1d.0ba2c9c ba NOM_INC_3 e8 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #7: 640x480 at 85 Hz → fpix = 36.0 MHz, N = 832 REGISTER NAME VALUE (DEC) VALUE (HEX) 832 340 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 40 03 NOM_INC_0 08 NOM_INC_1 e7 NOM_INC_2 25.454542 19.3a2e708 NOM_INC_3 a2 cb NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #8: 800x600 at 56.25 Hz → fpix = 36.0 MHz, N = 1024 REGISTER NAME VALUE (DEC) VALUE (HEX) 1024 400 TERM_CNT_0 TERM_CNT_1 00 NOM_INC_0 04 08 NOM_INC_1 NOM_INC_2 REGISTER VALUE (HEX) e7 25.454542 19.3a2e708 NOM_INC_3 a2 cb NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 A–3 Mode #9: 800x600 at 60 Hz → fpix = 40.0 MHz, N = 1056 REGISTER NAME VALUE (DEC) VALUE (HEX) 1056 420 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 20 04 NOM_INC_0 ee NOM_INC_1 cf NOM_INC_2 22.909088 16.745cfee 45 NOM_INC_3 b7 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #10: 800x600 at 72 Hz → fpix = 50.0 MHz, N = 1040 REGISTER NAME VALUE (DEC) VALUE (HEX) 1040 410 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HE10X) 10 04 NOM_INC_0 f2 NOM_INC_1 3f NOM_INC_2 18.327270 12.29e3ff2 NOM_INC_3 9e 92 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #11: 800x600 at 75 Hz → fpix = 49.5 MHz, N = 1056 REGISTER NAME VALUE (DEC) VALUE (HEX) 1056 420 TERM_CNT_0 TERM_CNT_1 20 NOM_INC_0 62 18.512394 12.4196235 NOM_INC_3 19 94 NOM_INC_4 A–4 04 35 NOM_INC_1 NOM_INC_2 REGISTER VALUE (HEX) 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #12: 800x600 at 85 Hz → fpix = 56.25 MHz, N=1048 REGISTER NAME VALUE (DEC) VALUE (HEX) 1048 418 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 18 04 NOM_INC_0 10 NOM_INC_1 c7 NOM_INC_2 16.290907 10.253c710 53 NOM_INC_3 82 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #13: 1024x768 at 43.4 Hz → fpix = 44.9 MHz, N = 1264 REGISTER NAME VALUE (DEC) VALUE (HEX) 1264 4f0 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) f0 04 NOM_INC_0 05 NOM_INC_1 9b NOM_INC_2 20.408987 14.3459b05 NOM_INC_3 45 a3 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 4 4 02 Mode #14: 1024x768 at 60 Hz → fpix = 65.0 MHz, N = 1344 REGISTER NAME VALUE (DEC) VALUE (HEX) 1344 540 TERM_CNT_0 TERM_CNT_1 40 NOM_INC_0 05 ea NOM_INC_1 NOM_INC_2 REGISTER VALUE (HEX) ff 28.195801 1c.190ffea NOM_INC_3 90 e1 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 2 2 01 A–5 Mode #15: 1024x768 at 70 Hz → fpix = 75.0 MHz, N = 1328 REGISTER NAME VALUE (DEC) VALUE (HEX) 1328 530 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 30 05 NOM_INC_0 97 NOM_INC_1 aa NOM_INC_2 24.436361 18.37daa97 7d NOM_INC_3 c3 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 2 2 01 Mode #16: 1024x768 at 75 Hz → fpix = 78.75 MHz, N = 1312 REGISTER NAME VALUE (DEC) VALUE (HEX) 1312 520 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 20 05 NOM_INC_0 16 NOM_INC_1 8a NOM_INC_2 23.272724 17.22e8a16 NOM_INC_3 2e ba NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 2 2 01 Mode #17: 1024x768 at 85 Hz → fpix = 94.5 MHz, N = 1376 REGISTER NAME VALUE (DEC) VALUE (HEX) 1376 560 TERM_CNT_0 TERM_CNT_1 60 NOM_INC_0 c8 19.393937 13.326c868 NOM_INC_3 26 9b NOM_INC_4 A–6 05 68 NOM_INC_1 NOM_INC_2 REGISTER VALUE (HEX) 0 VCO_DIV 8 8 03 SEL_CLK 2 2 01 DTV MODES 1080I → fpix = 74.25 MHz, N = 2200 REGISTER NAME VALUE (DEC) VALUE (HEX) 2200 898 TERM_CNT_0 TERM_CNT_1 98 NOM_INC_0 08 9b NOM_INC_1 NOM_INC_2 REGISTER VALUE (HEX) 2d 24.683192 18.5772d9b NOM_INC_3 77 c5 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 2 2 01 720P → fpix = 74.25 MHz, N = 1650 REGISTER NAME VALUE (DEC) VALUE (HEX) 1650 672 TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 72 06 NOM_INC_0 9b NOM_INC_1 2d NOM_INC_2 24.683192 18.5772d9b NOM_INC_3 77 c5 NOM_INC_4 0 VCO_DIV 8 8 03 SEL_CLK 2 2 01 VALUE (DEC) VALUE (HEX) REGISTER VALUE (HEX) 864 360 13.5 MHz, 864 REGISTER NAME TERM_CNT_0 TERM_CNT_1 60 03 NOM_INC_0 df NOM_INC_1 62 NOM_INC_2 29.696966 1d.59362df 93 NOM_INC_3 ed NOM_INC_4 0 VCO_DIV 7 7 02 SEL_CLK 8 8 03 A–7 13.5 MHz, N = 858 REGISTER NAME VALUE (DEC) VALUE (HEX) 858 35a TERM_CNT_0 TERM_CNT_1 REGISTER VALUE (HEX) 5a 03 NOM_INC_0 df NOM_INC_1 62 NOM_INC_2 29.696966 1d.59362df 93 NOM_INC_3 ed NOM_INC_4 0 VCO_DIV 7 7 02 SEL_CLK 8 8 03 VALUE (DEC) VALUE (HEX) REGISTER VALUE (HEX) 858 35a 14.31818 MHz, N = 858 REGISTER NAME TERM_CNT_0 TERM_CNT_1 5a 00 NOM_INC_1 00 NOM_INC_2 28.000000 1c.0000000 NOM_INC_3 00 E0 NOM_INC_4 A–8 03 NOM_INC_0 0 VCO_DIV 7 7 02 SEL_CLK 8 8 03