a FEATURES Fast 16-Bit ADC 100 kSPS Throughput Rate—AD977 200 kSPS Throughput Rate—AD977A Single 5 V Supply Operation Power Dissipation 100 mW Max Power-Down Mode 50 W Input Ranges: Unipolar; 0 V–10 V, 0 V–5 V and 0 V–4 V Bipolar; 10 V, 5 V and 3.3 V Choice of External or Internal 2.5 V Reference High Speed Serial Interface On-Chip Clock 20-Lead Skinny DIP or SOIC Package 28-Lead Skinny SSOP Package 16-Bit, 100 kSPS/200 kSPS BiCMOS A/D Converter AD977/AD977A FUNCTIONAL BLOCK DIAGRAM VANA REF 4k AD977/ AD977A 2.5V REFERENCE CAP 4R AGND1 R1IN R2IN R3IN AGND2 VDIG SYNC 2R R SWITCHED CAP ADC 4R SERIAL DATA INTERFACE BUSY DATACLK DATA R = 5k AD977 R = 2.5k AD977A CONTROL LOGIC & INTERNAL CALIBRATION CIRCUITRY DGND PWRD R/C CS TAG CLOCK SB/BTC EXT/INT GENERAL DESCRIPTION PRODUCT HIGHLIGHTS The AD977/AD977A is a high speed, low power 16-bit A/D converter that operates from a single 5 V supply. The AD977A has a throughput rate of 200 kSPS whereas the AD977 has a throughput rate of 100 kSPS. Each part contains a successive approximation, switched capacitor ADC, an internal 2.5 V reference, and a high speed serial interface. The ADC is factory calibrated to minimize all linearity errors. The AD977/AD977A is specified for full scale bipolar input ranges of ± 10 V, ± 5 V and ± 3.3 V, and unipolar ranges of 0 V to 10 V, 0 V to 5 V and 0 V to 4 V. 1. Fast Throughput The AD977/AD977A is a high speed, 16-bit ADC based on a factory calibrated switched capacitor architecture. The AD977/AD977A is comprehensively tested for ac parameters such as SNR and THD, as well as the more traditional dc parameters of offset, gain and linearity. 2. Single-Supply Operation The AD977/AD977A operates from a single 5 V supply and dissipates only 100 mW max. 3. Comprehensive DC and AC Specifications In addition to the traditional specifications of offset, gain and linearity, the AD977/AD977A is fully tested for SNR and THD. REV. D Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices. One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 781/329-4700 World Wide Web Site: http://www.analog.com Fax: 781/326-8703 © Analog Devices, Inc., 2000 AD977/AD977A AD977–SPECIFICATIONS (–40C to +85C, F = 100 kHz, V S Parameter Min RESOLUTION 16 ANALOG INPUT Voltage Range Impedance Sampling Capacitance THROUGHPUT SPEED Complete Cycle Throughput Rate DC ACCURACY Integral Linearity Error Differential Linearity Error No Missing Codes Transition Noise2 Full-Scale Error3, 4 Full-Scale Error Drift Full-Scale Error Ext. REF = 2.5 V Full-Scale Error Drift Ext. REF = 2.5 V Bipolar Zero Error3 Bipolar Ranges Bipolar Zero Error Drift Bipolar Ranges Unipolar Zero Error3 Unipolar Ranges Unipolar Zero Error Drift Unipolar Ranges Recovery to Rated Accuracy After Power-Down5 2.2 µF to CAP Power Supply Sensitivity VANA = VDIG = VD = 5 V ± 5% AC ACCURACY Spurious Free Dynamic Range6 Total Harmonic Distortion6 Signal-to-(Noise+Distortion)6 –60 dB Input Signal-to-Noise6 Full Power Bandwidth8 –3 dB Input Bandwidth Max Min B Grade Typ = VANA = 5 V, unless otherwise noted) Max 16 ±3 +3 40 1.0 10 ± 2.0 +1.75 ±3 ±2 15 1.0 1.0 ± 0.5 ±7 ± 0.5 ± 0.25 ±7 ± 0.25 ±2 ± 10 ±2 ± 10 ±2 ± 10 µs kHz LSB1 LSB Bits LSB % ppm/°C % ppm/°C ± 15 ±2 ± 10 ± 0.5 ± 0.5 ±2 ±2 pF 100 –1 16 Unit Bits 10 100 ±7 C Grade Typ Max 16 10 100 –2 15 Min ± 10 V, 0 V to 5 V, . . . (See Table II) See Table II 40 40 mV ppm/°C ± 10 mV ±2 ±2 ±2 ppm/°C 1 1 1 ms ±8 ±8 90 96 ±8 90 –90 –96 83 85 27 –90 83 28 83 27 85 700 1.5 SAMPLING DYNAMICS Aperture Delay Transient Response, Full-Scale Step Overvoltage Recovery9 REFERENCE Internal Reference Voltage Internal Reference Source Current External Reference Voltage Range for Specified Linearity External Reference Current Drain Ext. REF = 2.5 V A Grade Typ DIG 83 700 1.5 40 700 1.5 40 2 40 2 150 2 150 150 LSB dB7 dB dB dB dB kHz MHz ns µs ns 2.48 2.5 1 2.52 2.48 2.5 1 2.52 2.48 2.5 1 2.52 V µA 2.3 2.5 2.7 2.3 2.5 2.7 2.3 2.5 2.7 V 100 µA 100 100 NOTES 1 LSB means Least Significant Bit. With a ± 10 V input, one LSB is 305 µV. 2 Typical rms noise at worst case transitions and temperatures. 3 Measured with fixed resistors as shown in Figures 11, 12 and 13. Adjustable to zero. Tested at room temperature. 4 Full-Scale Error is expressed as the % difference between the actual full-scale code transition voltage and the ideal full scale transition voltage, and includes the effect of offset error. For bipolar input ranges, the Full-Scale Error is the worst case of either the –Full Scale or +Full Scale code transition voltage errors. For unipolar input ranges, Full-Scale Error is with respect to the +Full-Scale code transition voltage. 5 External 2.5 V reference connected to REF. 6 fIN = 20 kHz, 0.5 dB down unless otherwise noted. 7 All specifications in dB are referred to a full scale ± 10 V input. 8 Full-Power Bandwidth is defined as full-scale input frequency at which Signal-to-(Noise+Distortion) degrades to 60 dB, or 10 bits of accuracy. 9 Recovers to specified performance after a 2 × FS input overvoltage. Specifications subject to change without notice. –2– REV. D AD977/AD977A AD977A–SPECIFICATIONS (–40C to +85C, F = 200 kHz, V S Parameter Min RESOLUTION 16 ANALOG INPUT Voltage Range Impedance Sampling Capacitance THROUGHPUT SPEED Complete Cycle Throughput Rate DC ACCURACY Integral Linearity Error Differential Linearity Error No Missing Codes Transition Noise2 Full-Scale Error3, 4 Full-Scale Error Drift Full-Scale Error Ext. REF = 2.5 V Full-Scale Error Drift Ext. REF = 2.5 V Bipolar Zero Error3 Bipolar Ranges Bipolar Zero Error Drift Bipolar Ranges Unipolar Zero Error3 Unipolar Ranges Unipolar Zero Error Drift Unipolar Ranges Recovery to Rated Accuracy After Power-Down5 2.2 µF to CAP Power Supply Sensitivity VANA = VDIG = VD = 5 V ± 5% AC ACCURACY Spurious Free Dynamic Range6 Total Harmonic Distortion6 Signal-to-(Noise+Distortion)6 –60 dB Input Signal-to-Noise6 Full Power Bandwidth8 –3 dB Input Bandwidth Max Min B Grade Typ = VANA = 5 V, unless otherwise noted) Max 16 ±3 +3 40 1.0 5 ± 2.0 +1.75 ±3 ±2 15 1.0 1.0 ± 0.5 ±7 ± 0.5 ± 0.25 ±7 ± 0.25 ±2 ± 10 ±2 ± 10 ±2 ± 10 µs kHz LSB1 LSB Bits LSB % ppm/°C % ppm/°C ± 15 ±2 ± 10 ± 0.5 ± 0.5 ±2 ±2 pF 200 –1 16 Unit Bits 5 200 ±7 C Grade Typ Max 16 5 200 –2 15 Min ± 10 V, 0 V to 5 V, . . . (See Table II) See Table II 40 40 mV ppm/°C ± 10 mV ±2 ±2 ±2 ppm/°C 1 1 1 ms ±8 ±8 90 96 ±8 90 –90 –96 83 85 27 –90 83 28 83 27 85 1 2.7 SAMPLING DYNAMICS Aperture Delay Transient Response, Full-Scale Step Overvoltage Recovery9 REFERENCE Internal Reference Voltage Internal Reference Source Current External Reference Voltage Range for Specified Linearity External Reference Current Drain Ext. REF = 2.5 V A Grade Typ DIG 83 1 2.7 40 1 2.7 40 1 40 1 150 1 150 150 LSB dB7 dB dB dB dB MHz MHz ns µs ns 2.48 2.5 1 2.52 2.48 2.5 1 2.52 2.48 2.5 1 2.52 V µA 2.3 2.5 2.7 2.3 2.5 2.7 2.3 2.5 2.7 V 1.2 mA 1.2 1.2 NOTES 1 LSB means Least Significant Bit. With a ± 10 V input, one LSB is 305 µV. 2 Typical rms noise at worst case transitions and temperatures. 3 Measured with fixed resistors as shown in Figures 11, 12 and 13. Adjustable to zero. Tested at room temperature. 4 Full-Scale Error is expressed as the % difference between the actual full-scale code transition voltage and the ideal full scale transition voltage, and includes the effect of offset error. For bipolar input ranges, the Full-Scale Error is the worst case of either the –Full Scale or +Full Scale code transition voltage errors. For unipolar input ranges, Full-Scale Error is with respect to the +Full-Scale code transition voltage. 5 External 2.5 V reference connected to REF. 6 fIN = 20 kHz, 0.5 dB down unless otherwise noted. 7 All specifications in dB are referred to a full scale ± 10 V input. 8 Full-Power Bandwidth is defined as full-scale input frequency at which Signal-to-(Noise+Distortion) degrades to 60 dB, or 10 bits of accuracy. 9 Recovers to specified performance after a 2 × FS input overvoltage. Specifications subject to change without notice. REV. D –3– AD977/AD977A–SPECIFICATIONS (Both Specs) Parameter Conditions DIGITAL INPUTS Logic Levels VIL VIH IIL IIH DIGITAL OUTPUTS Data Format Data Coding Pipeline Delay VOL VOH –0.3 2.0 ISINK = 1.6 mA ISOURCE = 500 µA POWER SUPPLIES Specified Performance VDIG VANA IDIG IANA Power Dissipation PWRD LOW PWRD HIGH TEMPERATURE RANGE Specified Performance A, B, C Grades Typ Min Max Unit +0.8 VDIG + 0.3 ± 10 ± 10 V V µA µA Serial 16-Bits Binary Two’s Complement or Straight Binary Conversion Results Only Available after Completed Conversion 0.4 4 4.75 4.75 5 5 4 11 5.25 5.25 V V mA mA 100 mW µW +85 °C 50 TMIN to TMAX –40 V V Specifications subject to change without notice. TIMING SPECIFICATIONS (AD977A: F = 200 kHz, AD977: F = 100 kHz, V S Convert Pulsewidth R/C, CS to BUSY Delay BUSY LOW Time BUSY Delay after End of Conversion Aperture Delay Conversion Time Acquisition Time Throughput Time R/C Low to DATACLK Delay DATACLK Period DATA Valid Setup Time DATA Valid Hold Time EXT. DATACLK Period EXT. DATACLK HIGH EXT. DATACLK LOW R/C, CS to EXT. DATACLK Setup Time R/C to CS Setup Time EXT. DATACLK to SYNC Delay EXT. DATACLK to DATA Valid Delay CS to EXT. DATACLK Rising Edge Delay Previous DATA Valid after CS, R/C Low BUSY to EXT. DATACLK Setup Time Final EXT. DATACLK to BUSY Rising Edge TAG Valid Setup Time TAG Valid Hold Time S Symbol Min t1 t2 t3 t4 t5 t6 t7 t 6 + t7 t8 t9 t10 t11 t12 t13 t14 t15 t16 t17 t18 t19 t20 t21 t22 t23 t24 50 DIG = VANA = 5 V, –40C to +85C) AD977A Typ Max Min AD977 Typ Max 50 83 4.0 50 40 3.8 83 8.0 50 40 7.6 4.0 1.0 5 10 220 220 50 20 66 20 30 20 10 15 25 10 3.5 5 350 450 t12 + 5 66 66 100 20 100 20 30 20 10 15 25 10 7.5 5 1.7 0 20 8.0 2.0 t12 + 5 66 66 3.5 0 20 Unit ns ns µs ns ns µs µs µs ns ns ns ns ns ns ns ns ns ns ns ns µs ns µs ns ns Specifications subject to change without notice. –4– REV. D AD977/AD977A PIN CONFIGURATIONS ABSOLUTE MAXIMUM RATINGS 1 Analog Inputs R1IN, R2IN , R3IN . . . . . . . . . . . . . . . . . . . . . . . . . . . . ± 25 V CAP . . . . . . . . . . . . . . . . .+VANA + 0.3 V to AGND2 – 0.3 V REF . . . . . . . . . . . . . . . . . . . . . Indefinite Short to AGND2, . . . . . . . . . . . . . . . . . . . . . . . . . Momentary Short to VANA Ground Voltage Differences DGND, AGND1, AGND2 . . . . . . . . . . . . . . . . . . . ± 0.3 V Supply Voltages VANA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 V VDIG to VANA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ± 7 V VDIG . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 V Digital Inputs . . . . . . . . . . . . . . . . . . . –0.3 V to VDIG + 0.3 V Internal Power Dissipation2 PDIP (N), SOIC (R), SSOP (RS) . . . . . . . . . . . . . 700 mW Junction Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . 150°C Storage Temperature Range N, R . . . . . . . . –65°C to +150°C Lead Temperature Range (Soldering 10 sec) . . . . . . . . . . . . . . . . . . . . . . . . . . . 300°C SOIC and DIP 20 VDIG R1IN 1 R2IN 3 CAP 5 R1IN 1 AD977 AD977A 27 VANA 18 PWRD R2IN 3 26 PWRD 17 BUSY R3IN 4 25 BUSY NC 5 16 CS CAP 6 TOP VIEW 15 REF 6 R/C (Not to Scale) 14 TAG AGND2 7 REF 7 24 CS 23 NC AD977 AD977A 22 NC TOP VIEW 21 R/C (Not to Scale) 20 NC AGND2 9 NC 8 SB/BTC 8 13 DATA EXT/INT 9 12 DATACLK DGND 10 28 VDIG AGND1 2 19 VANA AGND1 2 R3IN 4 SSOP 11 SYNC NC 10 19 TAG NC 11 18 NC SB/BTC 12 17 DATA EXT/INT 13 16 DATACLK 15 SYNC DGND 14 NC = NO CONNECT NOTES 1 Stresses above those listed under Absolute Maximum Ratings may cause permanent damage to the device. This is a stress rating only; functional operation of the device at these or any other conditions above those indicated in the operational section of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability. 2 Specification is for device in free air: 20-Lead PDIP: θJA = 100°C/W, θJC = 31°C/W, 20-Lead SOIC: θJA = 75°C/W, θJC = 24°C/W, 28-Lead SSOP: θJA = 109°C/W, θJC = 39°C/W. 1.6mA TO OUTPUT PIN IOL 1.4V CL 100pF 500A IOH Figure 1. Load Circuit for Digital Interface Timing ORDERING GUIDE Model Temperature Range Throughput Rate AD977AN AD977BN AD977CN AD977AAN AD977ABN AD977ACN AD977AR AD977BR AD977CR AD977AAR AD977ABR AD977ACR AD977ARS AD977BRS AD977CRS AD977AARS AD977ABRS AD977ACRS –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C –40°C to +85°C 100 kSPS 100 kSPS 100 kSPS 200 kSPS 200 kSPS 200 kSPS 100 kSPS 100 kSPS 100 kSPS 200 kSPS 200 kSPS 200 kSPS 100 kSPS 100 kSPS 100 kSPS 200 kSPS 200 kSPS 200 kSPS Max INL Min S/(N+D) Package Options* ± 3.0 LSB ± 2.0 LSB 83 dB 85 dB 83 dB 83 dB 85 dB 83 dB 83 dB 85 dB 83 dB 83 dB 85 dB 83 dB 83 dB 85 dB 83 dB 83 dB 85 dB 83 dB N-20 N-20 N-20 N-20 N-20 N-20 R-20 R-20 R-20 R-20 R-20 R-20 RS-28 RS-28 RS-28 RS-28 RS-28 RS-28 ± 3.0 LSB ± 2.0 LSB ± 3.0 LSB ± 2.0 LSB ± 3.0 LSB ± 2.0 LSB ± 3.0 LSB ± 2.0 LSB ± 3.0 LSB ± 2.0 LSB *N = 20-lead 300 mil plastic DIP; R = 20-lead SOIC; RS = 28-lead SSOP. CAUTION ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily accumulate on the human body and test equipment and can discharge without detection. Although the AD977/AD977A feature proprietary ESD protection circuitry, permanent damage may occur on devices subjected to high-energy electrostatic discharges. Therefore, proper ESD precautions are recommended to avoid performance degradation or loss of functionality. REV. D –5– WARNING! ESD SENSITIVE DEVICE AD977/AD977A PIN FUNCTION DESCRIPTIONS Pin No. DIP/SOIC Pin No. SSOP Mnemonic Description 1, 3, 4 2 5 1, 3, 4 2 6 R1IN, R2IN, R3IN AGND1 CAP 6 7 REF 7 8 9 12 AGND2 SB/BTC 9 13 EXT/INT 10 11 14 15 DGND SYNC 12 16 DATACLK 13 17 DATA 14 19 TAG 15 21 R/C 16 24 CS 17 25 BUSY 18 26 PWRD 19 20 27 28 VANA VDIG Analog Input. Refer to Table I, Table II for input range configuration. Analog Ground. Used as the ground reference point for the REF pin. Reference buffer output. Connect a 2.2 µF tantalum capacitor between CAP and Analog Ground. Reference Input/Output. The internal 2.5 V reference is available at this pin. Alternatively an external reference can be used to override the internal reference. In either case, connect a 2.2 µF tantalum capacitor between REF and Analog Ground. Analog Ground. This digital input is used to select the data format of a conversion result. With SB/BTC tied LOW, conversion data will be output in Binary Two’s Complement format. With SB/BTC connected to a logic HIGH, data is output in Straight Binary format. Digital select input for choosing the internal or an external data clock. With EXT/INT tied LOW, after initiating a conversion, 16 DATACLK pulses transmit the previous conversion result as shown in Figure 3. With EXT/INT set to a logic HIGH, output data is synchronized to an external clock signal connected to the DATACLK input. Data is output as indicated in Figure 4 through Figure 9. Digital Ground. Digital output frame synchronization for use with an external data clock (EXT/INT = Logic HIGH). When a read sequence is initiated, a pulse one DATACLK period wide is output synchronous to the external data clock. Serial data clock input or output, dependent upon the logic state of the EXT/INT pin. When using the internal data clock (EXT/INT = Logic LOW), a conversion start sequence will initiate transmission of 16 DATACLK periods. Output data is synchronous to this clock and is valid on both its rising and falling edges (Figure 3). When using an external data clock (EXT/INT = Logic HIGH), the CS and R/C signals control how conversion data is accessed. The serial data output is synchronized to DATACLK. Conversion results are stored in an on-chip register. The AD977 provides the conversion result, MSB first, from its internal shift register. The DATA format is determined by the logic level of SB/BTC. When using the internal data clock (EXT/INT = Logic LOW), DATA is valid on both the rising and falling edges of DATACLK. Between conversions DATA will remain at the level of the TAG input when the conversion was started. Using an external data clock (EXT/INT = Logic HIGH) allows previous conversion data to be accessed during a conversion (Figures 5, 7 and 9) or the conversion result can be accessed after the completion of a conversion (Figures 4, 6 and 8). This digital input can be used with an external data clock, (EXT/INT = Logic HIGH) to daisy chain the conversion results from two or more AD977s onto a single DATA line. The digital data level on TAG is output on DATA with a delay of 16 or 17 external DATACLK periods after the initiation of the read sequence. Dependent on whether a SYNC is not present or present. Read/Convert Input. Is used to control the conversion and read modes of the AD977. With CS LOW; a falling edge on R/C holds the analog input signal internally and starts a conversion, a rising edge enables the transmission of the conversion result. Chip Select Input. With R/C LOW, a falling edge on CS will initiate a conversion. With R/C HIGH, a falling edge on CS will enable the serial data output sequence. Busy Output. Goes LOW when a conversion is started, and remains LOW until the conversion is completed and the data is latched into the on-chip shift register. Power-Down Input. When set to a logic HIGH power consumption is reduced and conversions are inhibited. The conversion result from the previous conversion is stored in the onboard shift register. Analog Power Supply. Nominally 5 V. Digital Power Supply. Nominally 5 V. –6– REV. D AD977/AD977A DEFINITION OF SPECIFICATIONS SPURIOUS FREE DYNAMIC RANGE INTEGRAL NONLINEARITY ERROR (INL) The difference, in decibels (dB), between the rms amplitude of the input signal and the peak spurious signal. Linearity error refers to the deviation of each individual code from a line drawn from “negative full scale” through “positive full scale.” The point used as “negative full scale” occurs 1/2 LSB before the first code transition. “Positive full scale” is defined as a level 1 1/2 LSB beyond the last code transition. The deviation is measured from the middle of each particular code to the true straight line. TOTAL HARMONIC DISTORTION (THD) THD is the ratio of the rms sum of the first six harmonic components to the rms value of a full-scale input signal and is expressed in decibels. SIGNAL TO (NOISE AND DISTORTION) (S/[N+D]) RATIO S/(N+D) is the ratio of the rms value of the measured input signal to the rms sum of all other spectral components below the Nyquist frequency, including harmonics but excluding dc. The value for S/(N+D) is expressed in decibels. DIFFERENTIAL NONLINEARITY ERROR (DNL) In an ideal ADC, code transitions are 1 LSB apart. Differential nonlinearity is the maximum deviation from this ideal value. It is often specified in terms of resolution for which no missing codes are guaranteed. FULL POWER BANDWIDTH The full power bandwidth is defined as the full-scale input frequency at which the S/(N+D) degrades to 60 dB, 10 bits of accuracy. FULL-SCALE ERROR The last + transition (from 011 . . . 10 to 011 . . . 11 for two’s complement format) should occur for an analog voltage 1 1/2 LSB below the nominal full scale (9.9995422 V for a ± 10 V range). The full-scale error is the deviation of the actual level of the last transition from the ideal level. APERTURE DELAY Aperture delay is a measure of the acquisition performance, and is measured from the falling edge of the R/C input to when the input signal is held for a conversion. BIPOLAR ZERO ERROR Bipolar zero error is the difference between the ideal midscale input voltage (0 V) and the actual voltage producing the midscale output code. TRANSIENT RESPONSE The time required for the AD977/AD977A to achieve its rated accuracy after a full-scale step function is applied to its input. UNIPOLAR ZERO ERROR OVERVOLTAGE RECOVERY In unipolar mode, the first transition should occur at a level 1/2 LSB above analog ground. Unipolar zero error is the deviation of the actual transition from that point. REV. D The time required for the ADC to recover to full accuracy after an analog input signal 150% of full-scale is reduced to 50% of the full-scale value. –7– AD977/AD977A CONVERSION CONTROL INTERNAL DATA CLOCK MODE The AD977/AD977A is controlled by two signals: R/C and CS. When R/C is brought low, with CS low, for a minimum of 50 ns, the input signal will be held on the internal capacitor array and a conversion “n” will begin. Once the conversion process does begin, the BUSY signal will go low until the conversion is complete. Internally, the signals R/C and CS are OR’d together and there is no requirement on which signal is taken low first when initiating a conversion. The only requirement is that there be at least 10 ns of delay between the two signals being taken low. After the conversion is complete the BUSY signal will return high and the AD977/AD977A will again resume tracking the input signal. Under certain conditions the CS pin can be tied Low and R/C will be used to determine whether you are initiating a conversion or reading data. On the first conversion, after the AD977/AD977A is powered up, the DATA output will be indeterminate. The AD977/AD977A is configured to generate and provide the data clock when the EXT/INT pin is held low. Typically CS will be tied low and R/C will be used to initiate a conversion “n.” During the conversion the AD977/AD977A will output 16 bits of data, MSB first, from conversion “n-1” on the DATA pin. This data will be synchronized with 16 clock pulses provided on the DATACLK pin. The output data will be valid on both the rising and falling edge of the data clock as shown in Figure 3. After the LSB has been presented, the DATA pin will assume whatever state the TAG input was at during the start of conversion, and the DATACLK pin will stay low until another conversion is initiated. EXTERNAL DATA CLOCK MODE The AD977/AD977A is configured to accept an externally supplied data clock when the EXT/INT pin is held high. This mode of operation provides several methods by which conversion results can be read from the AD977/AD977A. The output data from conversion “n-1” can be read during conversion “n,” or the output data from conversion “n” can be read after the conversion is complete. The external clock can be either a continuous or discontinuous clock. A discontinuous clock can be either Conversion results can be clocked serially out of the AD977/ AD977A using either an internal clock, generated by the AD977/AD977A, or by using an external clock. The AD977/ AD977A is configured for the internal data clock mode by pulling the EXT/INT pin low. It is configured for the external clock mode by pulling the EXT/INT pin high. t1 CS, R/C t3 BUSY t2 t4 t5 MODE ACQUIRE CONVERT ACQUIRE t6 t7 CONVERT Figure 2. Basic Conversion Timing t8 R/C t9 t1 DATACLK 1 t10 DATA 2 3 15 16 BIT 13 VALID BIT 1 VALID LSB VALID t11 MSB VALID BIT 14 VALID t2 t6 BUSY Figure 3. Serial Data Timing for Reading Previous Conversion Results with Internal Clock (CS, EXT/ INT and TAG Set to Logic Low) –8– REV. D AD977/AD977A normally low or normally high when inactive. In the case of the discontinuous clock, the AD977/AD977A can be configured to either generate or not generate a SYNC output (with a continuous clock a SYNC output will always be produced). EXTERNAL DISCONTINUOUS CLOCK DATA READ AFTER CONVERSION NO SYNC OUTPUT GENERATED Figure 4 illustrates the method by which data from conversion “n” can be read after the conversion is complete using a discontinuous external clock without the generation of a SYNC output. After a conversion is complete, indicated by BUSY returning high, the result of that conversion can be read while CS is Low and R/C is high. In this mode CS can be tied low. The MSB will be valid on the first falling edge and the second rising edge of DATACLK. The LSB will be valid on the 16th falling edge and the 17th rising edge of DATACLK. A minimum of 16 clock pulses are required for DATACLK if the receiving device will be latching data on the falling edge of DATACLK. A minimum of 17 clock pulses are required for DATACLK if the receiving device will be latching data on the rising edge of DATACLK. Approximately 40 ns after the 17th rising edge of DATACLK (if provided) the DATA output pin will reflect the state of the TAG input pin during the first rising edge of DATACLK. Each of the methods will be described in the following sections and are illustrated in Figures 4 through 9. It should be noted that all timing diagrams assume that the receiving device is latching data on the rising edge of the external clock. If the falling edge of DATACLK is used then, in the case of a discontinuous clock, one less clock pulse is required than shown in Figures 4 through 7 to latch in a 16-bit word. Note that data is valid on the falling edge of a clock pulse (for t13 greater than t18) and the rising edge of the next clock pulse. The AD977 provides error correction circuitry that can correct for an improper bit decision made during the first half of the conversion cycle. Normally the occurrence of an incorrect bit decision during a conversion cycle is irreversible. This error occurs as a result of noise during the time of the decision or due to insufficient settling time. As the AD977/AD977A is performing a conversion it is important that transitions not occur on digital input/output pins or degradation of the conversion result could occur. This is particularly important during the second half of the conversion process. For this reason it is recommended that when an external clock is being provided it be a discontinuous clock that is not toggling during the time that BUSY is low or, more importantly, that it does not transition during the latter half of BUSY low. t 13 EXT DATACLK 0 The advantage of this method of reading data is that it is not being clocked out during a conversion and therefore conversion performance is not degraded. When reading data after the conversion is complete, with the highest frequency permitted for DATACLK (15.15 MHz), and with the AD977A, the maximum possible throughput is approximately 195 kHz and not the rated 200 kHz. For details on use of the TAG input with this mode see the Use of the Tag Feature section. t 12 t 14 1 2 3 14 15 16 t1 R/C t2 BUSY t 21 SYNC t18 t18 DATA t 23 TAG BIT 15 (MSB) BIT 14 BIT 13 BIT 1 BIT 0 (LSB) TAG 1 TAG 2 TAG 3 TAG 15 TAG 16 TAG 0 TAG 1 t 24 TAG 0 TAG 17 TAG 18 Figure 4. Conversion and Read Timing Using an External Discontinuous Data Clock (EXT/ INT Set to Logic High, CS Set to Logic Low) REV. D –9– AD977/AD977A EXTERNAL DISCONTINUOUS CLOCK DATA READ DURING CONVERSION NO SYNC OUTPUT GENERATED EXTERNAL DISCONTINUOUS CLOCK DATA READ AFTER CONVERSION WITH SYNC OUTPUT GENERATED Figure 6 illustrates the method by which data from conversion “n” can be read after the conversion is complete using a discontinuous external clock, with the generation of a SYNC output. What permits the generation of a SYNC output is a transition of DATACLK while either CS is high or while both CS and R/C are low. After a conversion is complete, indicated by BUSY returning high, the result of that conversion can be read while CS is Low and R/C is high. In this mode CS can be tied low. In Figure 6 clock pulse #0 is used to enable the generation of a SYNC pulse. The SYNC pulse is actually clocked out approximately 40 ns after the rising edge of clock pulse #1. The SYNC pulse will be valid on the falling edge of clock pulse #1 and the rising edge of clock pulse #2. The MSB will be valid on the falling edge of clock pulse #2 and the rising edge of clock pulse #3. The LSB will be valid on the falling edge of clock pulse #17 and the rising edge of clock pulse #18. Approximately 40 ns after the rising edge of clock pulse #18 the DATA output pin will reflect the state of the TAG input pin during the rising edge of clock pulse #2. The advantage of this method of reading data is that it is not being clocked out during a conversion and therefore conversion performance is not degraded. Figure 5 illustrates the method by which data from conversion “n-1” can be read during conversion “n” while using a discontinuous external clock, without the generation of a SYNC output. After a conversion is initiated, indicated by BUSY going low, the result of the previous conversion can be read while CS is low and R/C is high. In this mode CS can be tied low. The MSB will be valid on the 1st falling edge and the 2nd rising edge of DATACLK. The LSB will be valid on the 16th falling edge and the 17th rising edge of DATACLK. A minimum of 16 clock pulses are required for DATACLK if the receiving device will be latching data on the falling edge of DATACLK. A minimum of 17 clock pulses are required for DATACLK if the receiving device will be latching data on the rising edge of DATACLK. Approximately 40 ns after the 17th rising edge of DATACLK (if provided) the DATA output pin will reflect the state of the TAG input pin during the first rising edge of DATACLK. For both the AD977 and the AD977A the data should be clocked out during the first half of BUSY so not to degrade conversion performance. For the AD977 this requires use of a 4.8 MHz DATACLK or greater with data being read out as soon as the conversion process begins. For the AD977A it requires use of a 10 MHz DATACLK or greater. When reading data after the conversion is complete, with the highest frequency permitted for DATACLK (15.15 MHz), and with the AD977A, the maximum possible throughput is approximately 195 kHz and not the rated 200 kHz. It is not recommended that data be shifted through the TAG input in this mode as it will certainly result in clocking of data during the second half of the conversion. For details on use of the TAG input with this mode see the Use of the TAG Input section. t 12 t 13 EXT DATACLK t 14 0 1 2 15 16 t 22 t15 R/C t1 t 20 BUSY t2 t 21 SYNC t18 DATA t18 BIT 15 (MSB) BIT 14 BIT 0 (LSB) Figure 5. Conversion and Read Timing for Reading Previous Conversion Results During A Conversion Using External Discontinuous Data Clock (EXT/ INT Set to Logic High, CS Set to Logic Low) –10– REV. D AD977/AD977A t 13 EXT DATACLK 0 t15 t 12 t 14 1 2 3 4 17 18 t15 t15 R/C t2 BUSY t 17 SYNC t12 t18 t 18 DATA t 23 BIT 14 BIT 0 (LSB) TAG 0 TAG 1 TAG 2 TAG 1 TAG 2 TAG 16 TAG 17 TAG 18 TAG 19 t 24 TAG 0 TAG BIT 15 (MSB) Figure 6. Conversion and Read Timing Using An External Discontinuous Data Clock (EXT/ INT Set to Logic High, CS Set to Logic Low) EXTERNAL DISCONTINUOUS CLOCK DATA READ DURING CONVERSION WITH SYNC OUTPUT GENERATED began. Figure 7 shows R/C then going high and after a delay of greater than 15 ns (t15) clock pulse #1 can be taken high to request the SYNC output. The SYNC output will appear approximately 40 ns after this rising edge and will be valid on the falling edge of clock pulse #1 and the rising edge of clock pulse #2. The MSB will be valid approximately 40 ns after the rising edge of clock pulse #2 and can be latched off either the falling edge of clock pulse #2 or the rising edge of clock pulse #3. The LSB will be valid on the falling edge of clock pulse #17 and the rising edge of clock pulse #18. Approximately 40 ns after the rising edge of clock pulse #18, the DATA output pin will reflect the state of the TAG input pin during the rising edge of clock pulse #2. Figure 7 illustrates the method by which data from conversion “n-1” can be read during conversion “n” while using a discontinuous external clock, with the generation of a SYNC output. What permits the generation of a SYNC output is a transition of DATACLK while either CS is High or while both CS and R/C are low. In Figure 7 a conversion is initiated by taking R/C low with CS tied low. While this condition exists a transition of DATACLK, clock pulse #0, will enable the generation of a SYNC pulse. Less then 83 ns after R/C is taken low the BUSY output will go low to indicate that the conversion process has t 12 t 13 EXT DATACLK 0 t15 t 14 1 2 3 18 t15 t 22 R/C t1 t 20 BUSY t2 t17 SYNC t 12 t18 DATA t18 BIT 15 (MSB) BIT 14 BIT 0 (LSB) TAG 0 Figure 7. Conversion and Read Timing for Reading Previous Conversion Results During a Conversion Using External Discontinuous Data Clock (EXT/ INT Set to Logic High, CS Set to Logic Low) REV. D –11– AD977/AD977A After a conversion is complete, indicated by BUSY returning high, the result of that conversion can be read while CS is low and R/C is high. In Figure 8 clock pulse #0 is used to enable the generation of a SYNC pulse. The SYNC pulse is actually clocked out approximately 40 ns after the rising edge of clock pulse #1. The SYNC pulse will be valid on the falling edge of clock pulse #1 and the rising edge of clock pulse #2. The MSB will be valid on the falling edge of clock pulse #2 and the rising edge of clock pulse #3. The LSB will be valid on the falling edge of clock pulse #17 and the rising edge of clock pulse #18. Approximately 50 ns after the rising edge of clock pulse #18 the DATA output pin will reflect the state of the TAG input pin during the rising edge of clock pulse #2. For both the AD977 and the AD977A the data should be clocked out during the first half of BUSY so not to degrade conversion performance. For the AD977 this requires use of a 4.8 MHz DATACLK or greater, with data being read out as soon as the conversion process begins. For the AD977A it requires use of a 10 MHz DATACLK or greater. It is not recommended that data be shifted through the TAG input in this mode as it will certainly result in clocking of data during the second half of the conversion. EXTERNAL CONTINUOUS CLOCK DATA READ AFTER CONVERSION WITH SYNC OUTPUT GENERATED Figure 8 illustrates the method by which data from conversion “n” can be read after the conversion is complete using a continuous external clock, with the generation of a SYNC output. What permits the generation of a SYNC output is a transition of DATACLK while either CS is high or while both CS and R/C are low. When reading data after the conversion is complete, with the highest frequency permitted for DATACLK (15.15 MHz) and, with the AD977A, the maximum possible throughput is approximately 195 kHz and not the rated 200 kHz. For details on use of the TAG input with this mode see the Use of the TAG Input section. With a continuous clock the CS pin cannot be tied low as it could be with a discontinuous clock. Use of a continuous clock, while a conversion is occurring, can increase the DNL and Transition Noise of the AD977/AD977A. t 13 EXT DATACLK t 12 t 14 0 1 t1 2 3 4 17 18 t19 t15 CS t 16 R/C t2 t16 BUSY t17 SYNC t12 t18 t18 DATA t23 TAG BIT 15 (MSB) BIT 14 BIT 0 (LSB) TAG 0 TAG 1 TAG 2 TAG 1 TAG 2 TAG 16 TAG 17 TAG 18 TAG 19 t24 TAG 0 Figure 8. Conversion and Read Timing Using an External Continuous Data Clock (EXT/ INT Set to Logic High) –12– REV. D AD977/AD977A With a continuous clock the CS pin cannot be tied low as it could be with a discontinuous clock. Use of a continuous clock while a conversion is occurring can increase the DNL and Transition Noise of the AD977/AD977A. 15 ns (t15), clock pulse #1 can be taken high to request the SYNC output. The SYNC output will appear approximately 50 ns after this rising edge and will be valid on the falling edge of clock pulse #1 and the rising edge of clock pulse #2. The MSB will be valid approximately 40 ns after the rising edge of clock pulse #2 and can be latched off either the falling edge of clock pulse #2 or the rising edge of clock pulse #3. The LSB will be valid on the falling edge of clock pulse #17 and the rising edge of clock pulse #18. Approximately 40 ns after the rising edge of clock pulse #18, the DATA output pin will reflect the state of the TAG input pin during the rising edge of clock pulse #2. In Figure 9 a conversion is initiated by taking R/C low with CS held low. While this condition exists a transition of DATACLK, clock pulse #0, will enable the generation of a SYNC pulse. Less then 83 ns after R/C is taken low the BUSY output will go low to indicate that the conversion process has began. Figure 9 shows R/C then going high and after a delay of greater than For both the AD977 and the AD977A the data should be clocked out during the 1st half of BUSY so as not to degrade conversion performance. For the AD977 this requires use of a 4.8 MHz DATACLK or greater with data being read out as soon as the conversion process begins. For the AD977A it requires use of a 10 MHz DATACLK or greater. EXTERNAL CONTINUOUS CLOCK DATA READ DURING CONVERSION WITH SYNC OUTPUT GENERATED Figure 9 illustrates the method by which data from conversion “n-1” can be read during conversion “n” while using a continuous external clock with the generation of a SYNC output. What permits the generation of a SYNC output is a transition of DATACLK while either CS is high or while both CS and R/C are low. t 12 t 13 t 14 EXT DATACLK 0 1 2 3 18 t 19 CS t 15 t 16 R/C t1 t 20 BUSY t2 t 17 SYNC t 12 t 18 t 18 BIT 15 (MSB) DATA t 23 TAG BIT 0 (LSB) TAG 0 TAG 16 TAG 17 TAG 1 TAG 2 t 24 TAG 0 TAG 1 TAG 18 TAG 19 Figure 9. Conversion and Read Timing for Reading Previous Conversion Results During a Conversion Using An External Continuous Data Clock (EXT/ INT Set to Logic High) REV. D –13– AD977/AD977A Table I. AD977A Analog Input Configuration Input Voltage Range Connect R1IN via 200 to Connect R2IN via 100 to Connect R3IN to Input Impedance ± 10 V ±5 V ± 3.3 V 0 V to 10 V 0 V to 5 V 0 V to 4 V VIN AGND VIN AGND AGND VIN AGND VIN VIN VIN AGND AGND 2.5 V 2.5 V 2.5 V AGND VIN VIN 11.5 kΩ 6.7 kΩ 5.4 kΩ 6.7 kΩ 5.0 kΩ 5.4 kΩ Table II. AD977 Analog Input Configuration Input Voltage Range Connect R1IN via 200 to Connect R2IN via 100 to Connect R3IN to Input Impedance ± 10 V ±5 V ± 3.3 V 0 V to 10 V 0 V to 5 V 0 V to 4 V VIN AGND VIN AGND AGND VIN AGND VIN VIN VIN AGND AGND CAP CAP CAP AGND VIN VIN 22.9 kΩ 13.3 kΩ 10.7 kΩ 13.3 kΩ 10.0 kΩ 10.7 kΩ ANALOG INPUTS The AD977/AD977A is specified to operate with six full-scale analog input ranges. Connections required for each of the three analog inputs, R1IN, R2IN and R3IN, and the resulting full-scale ranges, are shown in Table I and Table II. The nominal input impedance for each analog input range is also shown. Table III shows the output codes for the ideal input voltages of each of the six analog input ranges. The analog input section has a ± 25 V overvoltage protection on R1IN and R2IN. Since the AD977/AD977A has two analog grounds it is important to ensure that the analog input is referenced to the AGND1 pin, the low current ground. This will minimize any problems associated with a resistive ground drop. It is also important to ensure that the analog input of the AD977/AD977A is driven by a low impedance source. With its primarily resistive analog input circuitry, the ADC can be driven by a wide selection of general purpose amplifiers. To best match the low distortion requirements of the AD977/ AD977A, care should be taken in the selection of the drive circuitry op amp. that a single 5 V supply can be used to bias the hardware trim circuitry. With the hardware adjust circuits shown in Figures 12 and 13, offset and full-scale error can be trimmed to zero. Refer to the Offset and Gain Adjust section. If larger offset and full-scale errors are permitted, or if software calibration is used, the external resistors can be omitted. Table IV shows the resultant input ranges and offset and full-scale errors. Using the AD977A with Bipolar Input Ranges The connection diagrams in Figure 11 show a buffer amplifier required for bipolar operation of the AD977A when using the internal reference. The buffer amplifier is required to isolate the CAP pin from the signal dependent current in the R3IN pin. A high speed op amp such as the AD8031 can be used with a single 5 V power supply without degrading the performance of the AD977A. The buffer must have good settling characteristics and provide low total noise within the input bandwidth of the AD977A. REF AGND1 Figure 10 shows the simplified analog input section for the AD977/AD977A. Since the AD977/AD977A can operate with an internal or external reference, and several different analog input ranges, the full-scale analog input range is best represented with a voltage that spans 0 V to VREF across the 40 pF sampling capacitor. The onboard resistors are laser trimmed to ratio match for adjustment of offset and full-scale error using fixed external resistors. The configurations shown in Figures 12 and 13 are required to obtain the data sheet specifications for offset and full-scale error. The external fixed resistors are used during factory calibration so –14– 4k CAP 2.5V REFERENCE 20k /10k R1IN SWITCHED CAP ADC 10k /5k R2IN 5k /2.5k R3IN AGND2 20k /10k 40pF AD977/AD977A Figure 10. AD977/AD977A Simplified Analog Input REV. D AD977/AD977A Table III. Output Codes and Ideal Input Voltages Description Digital Output Two’s Complement Straight Binary (SB/BTC LOW) (SB/BTC HIGH) Analog Input Full-Scale Range Least Significant Bit +Full Scale (FS–1 LSB) Midscale One LSB Below Midscale –Full Scale ± 10 V 305 µV 9.999695 V 0V –305 µV –10 V ±5 V 153 µV 4.999847 V 0V –153 µV –5 V ± 3.33 V 102 µV 3.333231 V 0V –102 µV –3.333333 V 0 V to 10 V 153 µV 9.999847 V 5V 4.999847 V 0V 0 V to 5 V 76 µV 4.999924 V 2.5 V 2.499924 V 0V 0 V to 4 V 61 µV 3.999939 V 2V 1.999939 V 0V 0111 1111 1111 1111 0000 0000 0000 0000 1111 1111 1111 1111 1000 0000 0000 0000 1111 1111 1111 1111 1000 0000 0000 0000 0111 1111 1111 1111 0000 0000 0000 0000 Table IV. Input Ranges, Offset and Full-Scale Errors Without External Resistors AD977 Input Range Offset Error A/B/C Grade Full-Scale Error A/B/C Grade AD977A Input Range Offset Error A/B/C Grade Full-Scale Error A/B/C Grade –9.890 V to 9.90 V –4.943 V to 4.995 V –3.295 V to 3.330 V 0.008 V to 9.946 V 0.004 V to 5.023 V 0.003 V to 4.010 V ± 25 mV/±25 mV ± 25 mV/±25 mV ± 25 mV/±25 mV ± 10 mV/±10 mV ± 10 mV/±10 mV ± 10 mV/±10 mV ± 0.75%/± 0.50% ± 0.75%/± 0.50% ± 0.75%/± 0.50% ± 0.75%/± 0.50% ± 0.75%/± 0.50% ± 0.75%/± 0.50% –9.800 V to 9.970 V –4.900 V to 4.985 V –3.267 V to 3.323 V 0.007 V to 9.893 V 0.004 V to 5.039 V 0.003 V to 4.016 V ± 40 mV/±40 mV ± 40 mV/±40 mV ± 40 mV/±40 mV ± 10 mV/±10 mV ± 10 mV/±10 mV ± 10 mV/±10 mV ± 0.80%/± 0.55% ± 0.80%/± 0.55% ± 0.80%/± 0.55% ± 0.75%/± 0.50% ± 0.75%/± 0.50% ± 0.75%/± 0.50% 200 200 VIN 200 R1IN R1IN AGND1 AGND1 VIN R1IN AGND1 100 100 100 R2IN 33.2k R3IN VIN R2IN R3IN AD8031 CAP AD977A CAP CAP AD8031 2.2 F 2.2 F AD977A AD977A REF 2.2 F REF 2.2 F REF 2.2 F AGND2 a. R3IN 33.2k 33.2k AD8031 2.2 F R2IN AGND2 b. AGND2 c. Figure 11. AD977A Bipolar Input Configuration Using the Internal Reference; (a) VIN = ±10 V, (b) VIN = ±5 V, (c) VIN = ±3.33 V REV. D –15– AD977/AD977A BIPOLAR CONNECTION FOR AD977 INPUT RANGE STANDARD CONNECTION WITHOUT OFFSET AND GAIN ADJUST STANDARD CONNECTION WITH OFFSET AND GAIN ADJUST 200 200 VIN VIN R1IN R1IN AGND1 AGND1 100 100 R2IN R2IN 5V 33.2k 10V R3IN R3IN 33.2k CAP 50k CAP 2.2F 5V 2.2F AD977 AD977 576k REF 50k REF 2.2F 2.2F AGND2 AGND2 200 200 R1IN R1IN 5V AGND1 AGND1 33.2k 50k 100 VIN R2IN 100 33.2k 5V VIN R3IN R2IN R3IN CAP 2.2F CAP 5V AD977 2.2F AD977 576k REF REF 50k 2.2F 2.2F AGND2 AGND2 200 200 VIN VIN R1IN AGND1 100 R1IN AGND1 100 5V R2IN R2IN 33.2k 33.2k 3.33V R3IN 50k R3IN CAP CAP 2.2F 5V 2.2F AD977 AD977 576k REF 50k REF 2.2F 2.2F AGND2 AGND2 Figure 12. AD977 Bipolar Analog Input Configuration –16– REV. D AD977/AD977A UNIPOLAR CONNECTION FOR AD977A AND AD977 INPUT RANGE STANDARD CONNECTION WITHOUT OFFSET AND GAIN ADJUST STANDARD CONNECTION WITH OFFSET AND GAIN ADJUST 200 200 R1IN R1IN AGND1 AGND1 100 100 VIN VIN R2IN R2IN 5V R3IN 33.2k 0V–10V 5V AD977/ AD977A 576k 2.2F 2.2F AGND2 AGND2 200 200 R1IN R1IN 5V AGND1 50k R2IN 33.2k VIN AGND1 33.2k 100 0V–5V R3IN 100 R2IN VIN R3IN CAP CAP AD977/ AD977A 2.2F REF 5V + 2.2F – 576k 2.2F AGND2 AGND2 200 200 VIN VIN R1IN R1IN AGND1 AGND1 100 100 R2IN R2IN R3IN R3IN 5V CAP 33.2k 33.2k 2.2F 50k CAP 2.2F AD977/ AD977A REF AD977/ AD977A 5V 576k REF 50k 2.2F 2.2F AGND2 AGND2 Figure 13. AD977/AD977A Unipolar Analog Input Configuration REV. D AD977/ AD977A REF 50k 2.2F 0V–4V AD977/ AD977A REF 50k REF R3IN CAP 50k CAP 2.2F 2.2F 33.2k –17– AD977/AD977A VOLTAGE REFERENCE The AD977/AD977A has an on-chip temperature compensated bandgap voltage reference that is factory trimmed to 2.5 V ± 20 mV. The accuracy of the AD977/AD977A over the specified temperature ranges is dominated by the drift performance of the voltage reference. The on-chip voltage reference is lasertrimmed to provide a typical drift of 7 ppm/°C. This typical drift characteristic is shown in Figure 14, which is a plot of the change in reference voltage (in mV) versus the change in temperature—notice the plot is normalized for zero error at 25°C. If improved drift performance is required, an external reference such as the AD780 should be used to provide a drift as low as 3 ppm/°C. In order to simplify the drive requirements of the voltage reference (internal or external), an onboard reference buffer is provided. The output of this buffer is provided at the CAP pin and is available to the user; however, when externally loading the reference buffer, it is important to make sure that proper precautions are taken to minimize any degradation in the ADC’s performance. Figure 15 shows the load regulation of the reference buffer. Notice that this figure is also normalized so that there is zero error with no dc load. In the linear region, the output impedance at this point is typically 1 Ω. Because of this 1 Ω output impedance, it is important to minimize any ac or input dependent loads that will lead to increased distortion. Any dc loads will simply act as a gain error. Although the typical characteristic of Figure 15 shows that the AD977/AD977A is capable of driving loads greater than 15 mA, it is recommended that the steady state current not exceed 2 mA. Using an External Reference In addition to the on-chip reference, an external 2.5 V reference can be applied. When choosing an external reference for a 16-bit application, however, careful attention should be paid to noise and temperature drift. These critical specifications can have a significant effect on the ADC performance. Figures 16a and 16b show the AD977/AD977A used in bipolar mode with the AD780 voltage reference applied to the REF pin. It is important to note that in Figure 16a the R3IN pin is connected to the CAP pin whereas in Figure 16b the R3IN pin of the AD977A is returned to the output of the external reference. The AD780 is a bandgap reference that exhibits ultralow drift, low initial error and low output noise. In Figure 16b, the value for C1 is only applicable to applications using the AD780. In applications using a different external reference a different value for C1 may be required. For low power applications, the REF192 provides a low quiescent current, high accuracy and low temperature drift solution. 200 VIN R1IN 100 R2IN 1mV/DIV 33.2k R3IN 0.1F 3 TEMP VOUT 6 REF C1 2.2F AD780 2 V IN 5V GND 4 AGND1 C3 1F C4 0.1F AD977 VANA –55 25 DEGREES CELSIUS C2 2.2F 125 Figure 14. Reference Drift CAP AGND2 Figure 16a. AD780 External Reference to AD977 Configured for ± 10 V Input Range 200 VIN R1IN dV ON CAP PIN – 10mV/DIV 100 R2IN 33.2k 0.1F 3 TEMP VOUT 6 REF C1 330F* AD780 5V 2 V IN R3IN GND 4 AGND1 C3 1F C4 0.1F AD977A VANA *ESR AT 100kHz MUST BE LESS THAN 0.3. C2 RECOMMEND KEMET T495 SERIES OR 2.2F SANYO 6SA330M. SOURCE CAPABILITY SINK CAPABILITY LOAD CURRENT – 5mA/DIV Figure 15. CAP Pin Load Regulation CAP AGND2 Figure 16b. AD780 External Reference to AD977A Configured for ± 10 V Input Range –18– REV. D AD977/AD977A OFFSET AND GAIN ADJUSTMENT The AD977/AD977A is factory trimmed to minimize gain, offset and linearity errors. In some applications, where the analog input signal is required to meet the full dynamic range of the ADC, the gain and offset errors need to be externally trimmed to zero. Figures 12 and 13 show the required trim circuitry to correct for these offset and gain errors. Where adjustment is required, offset error must be corrected before gain error. To achieve this in the bipolar input configuration, trim the offset potentiometer with the input voltage set to 1/2 LSB below ground. Then adjust the potentiometer until the major carry transition is located between 1111 1111 1111 1111 and 0000 0000 0000 0000. To adjust the gain error, an analog signal should be input at either the first code transition (ADC negative full scale) or the last code transition (ADC positive full scale). Thus, to adjust for full-scale error, an input voltage of FS/2 – 3/2 LSBs can be applied to VIN, and the gain potentiometer should be adjusted until the output code flickers between the last positive code transition 0111 1111 1111 1111 and 0111 1111 1111 1110. Should the first code transition need adjusting, the trim procedure should consist of applying an analog input signal of –FS/2 + 1/2 LSB to the VIN input and adjusting the trim until the output code flickers between 1000 0000 0000 0000 and 1000 0000 0000 0001. AC PERFORMANCE The AD977/AD977A is fully specified and tested for dynamic performance specifications. The ac parameters are required for signal processing applications such as speech recognition and spectrum analysis. These applications require information on the ADC’s effect on the spectral content of the input signal. Hence, the parameters for which the AD977/AD977A is specified include S/(N+D), THD and Spurious Free Dynamic Range. These terms are discussed in greater detail in the following sections. the AD977/AD977A could be oversampled by a factor of 2/4. This would yield a 3/6 dB improvement in the effective SNR performance. DC PERFORMANCE The factory calibration scheme used for the AD977/AD977A compensates for bit weight errors that may exist in the capacitor array. The mismatch in capacitor values is adjusted (using the calibration coefficients) during a conversion resulting in excellent dc linearity performance. Figures 18, 19, 20, 21, 22 and 23, respectively, show typical INL, typical DNL, typical positive and negative INL and DNL distribution plots for the AD977/AD977A at 25°C. A histogram test is a statistical method for deriving an A/D converter’s differential nonlinearity. A ramp input is sampled by the ADC and a large number of conversions are taken at each voltage level, averaged then stored. The effect of averaging is to reduce the transition noise by 1/n. If 64 samples are averaged at each point, the effect of transition noise is reduced by a factor of 8, i.e., a transition noise of 0.8 LSBs rms is reduced to 0.1 LSBs rms. Theoretically the codes, during a test of DNL, would all be the same size and therefore have an equal number of occurrences. A code with an average number of occurrences would have a DNL of “0.” A code that is different from the average would have a DNL that was either greater or less than zero LSB. A DNL of –1 LSB indicates that there is a missing code present at the 16-bit level and that the ADC exhibits 15bit performance. 100% 2.0 1.5 1.0 As a general rule, it is recommended that the results from several conversions be averaged to reduce the effects of noise and thus improve parameters such as S/(N+D) and THD. The ac performance of the AD977/AD977A can be optimized by operating the ADC at its maximum sampling rate of 100 kHz/200 kHz and digitally filtering the resulting bit stream to the desired signal bandwidth. By distributing noise over a wider frequency range the noise density in the frequency band of interest can be reduced. For example, if the required input bandwidth is 50 kHz, LSB 0.5 0 –0.5 –1.0 –1.5 –2.0 0 5 10 15 20 25 30 35 40 45 OUTPUT CODE – K 50 55 60 66 50 55 60 66 Figure 18. INL Plot 0 –10 –20 –30 100% 2.0 1.5 1.0 –50 –60 0.5 –70 LSB AMPLITUDE – dB –40 5280 POINT FFT FSAMPLE = 200kHz FIN = 20kHz, 0dB SNRD = 86dB THD = –101dB –80 0 –90 –0.5 –100 –110 –1.0 –120 –1.5 –130 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 FREQUENCY – kHz –2.0 Figure 17. FFT Plot REV. D 0 5 10 15 20 25 30 35 40 45 OUTPUT CODE – K Figure 19. DNL Plot –19– 50 90 45 80 40 70 NUMBER OF UNITS NUMBER OF UNITS AD977/AD977A 35 30 25 20 15 60 50 40 30 20 10 10 0 0 0 0.2 0.3 0.4 0.5 0.6 0.8 0.9 1 1.1 1.2 1.4 1.5 1.6 1.7 1.8 2 2.1 2.2 2.3 2.4 2.6 2.7 2.8 2.9 –1.22 –1.17 –1.12 –1.07 –1.02 –0.97 –0.92 –0.87 –0.82 –0.78 –0.73 –0.68 –0.63 –0.58 –0.53 –0.48 –0.43 –0.38 –0.33 –0.28 –0.23 –0.18 –0.14 –0.09 –0.04 5 POSITIVE INL – LSB NEGATIVE DNL – LSB Figure 20. Typical Positive INL Distribution (999 Units) Figure 23. Typical Negative DNL Distribution (999 Units) 90 60 SNR+D (dB) FOR AD977A 80 SINAD (dB) FOR VIN = 0dB NUMBER OF UNITS 50 40 30 20 10 60 50 SNR+D (dB) FOR AD977 40 30 20 10 –2.5 –2.4 –2.3 –2.2 –2.1 –2.0 –1.9 –1.8 –1.7 –1.6 –1.5 –1.4 –1.3 –1.2 –1.1 –1.0 –0.9 –0.8 –0.7 –0.6 –0.5 –0.4 –0.3 –0.2 –0.1 0 70 1 10 100 INPUT SIGNAL FREQUENCY – kHz 1000 NEGATIVE INL – LSB Figure 24. S/(N+D) vs. Input Frequency 120 110 100 105 80 100 –90 95 –95 60 40 –80 SFDR THD 90 85 20 80 –75 0.02 0.10 0.19 0.27 0.36 0.44 0.52 0.61 0.69 0.77 0.86 0.94 1.02 1.11 1.19 1.27 1.36 1.44 1.53 1.61 1.69 1.78 1.86 1.94 2.03 0 –85 –100 –105 SNRD –50 –25 0 25 50 75 TEMPERATURE – C THD – dB SFDR, S/N+D – dB NUMBER OF UNITS Figure 21. Typical Negative INL Distribution (999 Units) 100 125 –110 150 POSITIVE DNL – LSB Figure 25. AC Parameters vs. Temperature Figure 22. Typical Positive DNL Distribution (999 Units) –20– REV. D AD977/AD977A DC CODE UNCERTAINTY Ideally, a fixed dc input should result in the same output code for repetitive conversions; however, as a consequence of unavoidable circuit noise within the wideband circuits of the ADC, a range of output codes may occur for a given input voltage. Thus, when a dc signal is applied to the AD977/AD977A input and 10,000 conversions are recorded, the result will be a distribution of codes as shown in Figure 26. This histogram shows a bell shaped curve consistent with the Gaussian nature of thermal noise. The histogram is approximately seven codes wide. The standard deviation of this Gaussian distribution results in a code transition noise of 1 LSB rms. to throw away the null bit if the sync input is toggled each time the null bit appears. If the application does not require simultaneous sampling, the null bit can be completely avoided by delaying the R/C signal of each upstream device by one clock cycle with respect to its immediate downstream device. This bit time delay can be accomplished through a D-type flip-flop that delays the R/C signal at its D-input by one cycle of the serial data clock that is at its clock input. AD977/AD977A #2 (UPSTREAM) AD977/AD977A #1 (DOWNSTREAM) TAG TAG 4000 3500 3000 DATA DATA CS CS R/C R/C DCLK DCLK DATA OUT 2500 DCLK IN R/ IN 2000 IN 1500 Figure 27. Two AD977/AD977A’s Utilizing Tag 1000 500 0 –3 –2 –1 0 1 2 3 4 Figure 26. Histogram of 10,000 Conversions of a DC Input USE OF THE TAG INPUT The AD977/AD977A provides a TAG input pin for cascading multiple converters together. This feature is useful for reducing component count in systems where an isolation barrier must be crossed and is also useful for systems with a limited capacity for interfacing to a large number of converters. The tag feature only works in the external clock mode and requires that the DATA output of a “upstream” device be connected to the TAG input of an “downstream” device. An example of the concatenation of two devices is shown in Figure 27 and their resultant output is shown in Figure 28. In Figure 27, the paralleled R/C ensures that each AD977/ AD977A will simultaneously sample their inputs. In Figure 28, a “null” bit is shown between each 16-bit word associated with each ADC in the serial data output stream. This is the result of a minimum value for “External Data Clock to Data Valid Delay” (t18) that is greater than the “TAG Valid Setup Time” (t23). In other words, when you concatenate two or more AD977/AD977As the MSB on the downstream device will not be present on the TAG input of the upstream device in time to meet the setup time requirement of the TAG input. If the serial data stream is going to a parallel port of a microprocessor that is also providing the serial data clock, then the microprocessor’s firmware can be written to “throw away” the null bit. If the serial data stream is going to a serial port then external “glue” logic will have to be added to make the interface work. If the serial port has a “sync” input then this can be used REV. D It is not recommended that the TAG feature be used with the read during convert mode because this will require data to be clocked out during the second half of the conversion process. It is recommended that the read after convert mode be used in an application that wants to take advantage of the TAG feature. To improve the data throughput a combination of the two data read methods can be used and is described as follows. If two or more AD977/AD977As are to have their data output concatenated together in a single data stream, and if data throughput is to be maximized, a system could be designed such that the upstream device data is read during the first half of its conversion process and the remainder of the downstream devices read during the time between conversions. Assume three AD977As are to have their data concatenated. Assume the further most downstream device is referred to as device #1 and the further most upstream device as #3. Each device is driven from a common DATACLK and R/C control signal, the CS input of each device is tied to ground. The three BUSY outputs should be OR’d together to form a composite BUSY. After the conversion is complete, as indicated by the composite BUSY going high, an external, normally low, 15.15 MHz DATACLK can be toggled 34 times to first read the data first from device #3 and then from device #2. When the composite BUSY goes low to indicate the beginning of the conversion process the external DATACLK can be toggled 17 times to read the data from device #1 during the first half of the conversion process. Using this technique it would be possible to read in the data from the three devices in approximately 6.4 µs for a throughput of approximately 156 kHz The receiving device would have to deal with the null bit between data from device #2 and #3. The receiving device would also have to be capable of starting and stopping the external DATACLK at the appropriate times. The TAG input, when unused, should always be tied either high or low and not be allowed to float. –21– AD977/AD977A MICROPROCESSOR INTERFACING R/C BUSY DATA The AD977/AD977A is ideally suited for traditional dc measurement applications supporting a microprocessor, and ac signal processing applications interfacing to a digital signal processor. The AD977/AD977A is designed to interface with a general purpose serial port or I/O ports on a microcontroller. A variety of external buffers can be used with the AD977/AD977A to prevent digital noise from coupling into the ADC. The following sections illustrate the use of the AD977/AD977A with an SPI equipped microcontroller and the ADSP-2181 signal processor. NULL BIT 15 0 DEVICE DATA #1 15 DEVICE DATA #2 DCLK Figure 28. TAG Timing Diagram for Two Concatenated AD977/AD977As POWER-DOWN FEATURE SPI Interface The AD977/AD977A has analog and reference power-down capability through the PWRD pin. When the PWRD pin is taken high, the power consumption drops from a maximum value of 100 mW to a typical value of 50 µW. When in the power-down mode the previous conversion results are still available in the internal registers and can be read out providing it has not already been shifted out. Figure 29 shows a general interface diagram between the AD977/AD977A and an SPI equipped microcontroller. This interface assumes that the convert pulses will originate from the microcontroller and that the AD977/AD977A will act as the slave device. The convert pulse could be initiated in response to an internal timer interrupt. The reading of output data, one byte at a time, if necessary, could be initiated in response to the endof-conversion signal (BUSY going high). When used with an external reference, connected to the REF pin and a 2.2 µF capacitor, connected to the CAP pin, the power up recovery time is typically 1 ms. This typical value of 1 ms for recovery time depends on how much charge has decayed from the external 2.2 µF capacitor on the CAP pin and assumes that it has decayed to zero. The 1 ms recovery time has been specified such that settling to 16-bits has been achieved. SDI DATA SCK DATACLK I/O PORT R/C IRQ BUSY +5V SPI When used with the internal reference, the dominant time constant for power-up recovery is determined by the external capacitor on the REF pin and the internal 4K impedance seen at that pin. An external 2.2 µF capacitor is recommended for the REF pin. AD977/ AD977A EXT/INT CS TAG Figure 29. AD977/AD977A to SPI Interface ADSP-2181 Interface CONSIDERATIONS WHEN USING MULTIPLEXED INPUTS Consideration must be given to the effect on A/D performance in applications that require the use of analog multiplexers or analog switches to interface multiple signals to the AD977/AD977A. The nonzero “on” resistance of a multiplexer or switch, at the input to the AD977/AD977A, will increase the system offset and gain error. As an example, consider the AD977 configured for an input voltage range of ± 10 V dc. For every 5 Ω of source impedance (in addition to the required external 200 Ω input resistor) an offset error of 1 LSB would be introduced and the positive gain error would increase by an added 0.00375% of full scale. This error, due to nonzero source impedance, can be corrected through a hardware or software system level calibration, but will only be valid at the temperature and input voltage present at the time of calibration. Another factor to consider is that most analog multiplexers and switches exhibit a nonlinear relationship between input signal level and on resistance. This will introduce added distortion products that will degrade THD, S/(N+D) and INL. For these reasons it is recommended that an appropriate buffer be used between the output of the multiplexer and the input of the AD977. When switching the input to the multiplexer, and subsequently the input to the AD977, it is recommended that the transition be made to occur either immediately after the current conversion is complete or shortly after the beginning of a conversion. Figure 30 shows an interface between the AD977/AD977A and the ADSP-2181 Digital Signal Processor. The AD977/AD977A is configured for the Internal Clock mode (EXT/INT = 0) and will therefore act as the master device. The convert command is shown generated from an external oscillator in order to provide a low jitter signal appropriate for both dc and ac measurements. Because the SPORT, within the ADSP-2181, will be seeing a discontinuous external clock, some steps are required to ensure that the serial port is properly synchronized to this clock during each data read operation. The recommended procedure to ensure this is as follows, – enable SPORT0 through the System Control register – set the SCLK Divide register to zero – setup PF0 and PF1 as outputs by setting bits 0 and 1 in PFTYPE – force RFS0 low through PF0. The Receive Frame Sync signal has been programmed active high – enable AD977/AD977A by forcing CS = 0 through PF1 – enable SPORT0 Receive Interrupt through the IMASK register – wait for at least one full conversion cycle of the AD977/AD977A and throw away the received data – disable the AD977/AD977A by forcing CS = 1 through PF1 – wait for a period of time equal to one conversion cycle – force RFS0 high through PF0 – enable the AD977/AD977A by forcing CS = 0 through PF1 –22– REV. D AD977/AD977A The ADSP-2181 SPORT0 will now remain synchronized to the external discontinuous clock for all subsequent conversions. DATA DR0 SCLK0 ADSP-2181 connected with the least resistance back to the power supply. AGND1 is the low current analog supply ground and should be the analog common for the external reference, input op amp drive circuitry and the input resistor divider circuit. By applying the inputs referenced to this ground, any ground variations will be offset and have a minimal effect on the resulting analog input to the ADC. The digital ground pin, DGND, is the reference point for all of the digital signals that control the AD977/AD977A. DATACLK OSCILLATOR PF1 RFS0 PF0 R/C AD977/ AD977A CS The AD977/AD977A can be powered with two separate power supplies or with a single analog supply. When the system digital supply is noisy, or fast switching digital signals are present, it is recommended to connect the analog supply to both the VANA and VDIG pins of the AD977/AD977A and the system supply to the remaining digital circuitry. With this configuration, AGND1, AGND2 and DGND should be connected back at the ADC. When there is significant bus activity on the digital output pins, the digital and analog supply pins on the ADC should be separated. This would eliminate any high speed digital noise from coupling back to the analog portion of the AD977/ AD977A. In this configuration, the digital ground pin DGND should be connected to the system digital ground and be separate from the AGND pins. EXT/INT TAG SPORT0 CNTRL REG = 0x300F Figure 30. AD977/AD977A to ADSP-2181 Interface POWER SUPPLIES AND DECOUPLING The AD977/AD977A has two power supply input pins. VANA and VDIG provide the supply voltages to the analog and digital portions, respectively. VANA is the 5 V supply for the on-chip analog circuitry, and VDIG is the 5 V supply for the on-chip digital circuitry. The AD977/AD977A is designed to be independent of power supply sequencing and thus free from supply voltage induced latchup. With high performance linear circuits, changes in the power supplies can result in undesired circuit performance. Optimally, well regulated power supplies should be chosen with less than 1% ripple. The ac output impedance of a power supply is a complex function of frequency and will generally increase with frequency. Thus, high frequency switching, such as that encountered with digital circuitry, requires the fast transient currents that most power supplies cannot adequately provide. Such a situation results in large voltage spikes on the supplies. To compensate for the finite ac output impedance of most supplies, charge “reserves” should be stored in bypass capacitors. This will effectively lower the supplies impedance presented to the AD977/AD977A VANA and VDIG pins and reduce the magnitude of these spikes. Decoupling capacitors, typically 0.1 µF, should be placed close to the power supply pins of the AD977/AD977A to minimize any inductance between the capacitors and the VANA and VDIG pins. The AD977/AD977A may be operated from a single 5 V supply. When separate supplies are used, however, it is beneficial to have larger capacitors, 10 µF, placed between the logic supply (VDIG) and digital common (DGND) and between the analog supply (VANA) and the analog common (AGND2). Additionally, 10 µF capacitors should be located in the vicinity of the ADC to further reduce low frequency ripple. In systems where the device will be subjected to harsh environmental noise, additional decoupling may be required. GROUNDING The AD977/AD977A has three ground pins; AGND1, AGND2 and DGND. The analog ground pins are the “high quality” ground reference points and should be connected to the system analog common. AGND2 is the ground to which most internal ADC analog signals are referenced. This ground is most susceptible to current induced voltage drops and thus must be REV. D BOARD LAYOUT Designing with high resolution data converters requires careful attention to board layout. Trace impedance is a significant issue. A 1.22 mA current through a 0.5 Ω trace will develop a voltage drop of 0.6 mV, which is 2 LSBs at the 16-bit level over the 20 volt full-scale range. Ground circuit impedances should be reduced as much as possible since any ground potential differences between the signal source and the ADC appear as an error voltage in series with the input signal. In addition to ground drops, inductive and capacitive coupling needs to be considered. This is especially true when high accuracy analog input signals share the same board with digital signals. Thus, to minimize input noise coupling, the input signal leads to VIN and the signal return leads from AGND should be kept as short as possible. In addition, power supplies should also be decoupled to filter out ac noise. Analog and digital signals should not share a common path. Each signal should have an appropriate analog or digital return routed close to it. Using this approach, signal loops enclose a small area, minimizing the inductive coupling of noise. Wide PC tracks, large gauge wire and ground planes are highly recommended to provide low impedance signal paths. Separate analog and digital ground planes are also recommended with a single interconnection point to minimize ground loops. Analog signals should be routed as far as possible from high speed digital signals and should only cross them, if absolutely necessary, at right angles. In addition, it is recommended that multilayer PC boards be used with separate power and ground planes. When designing the separate sections, careful attention should be paid to the layout. –23– AD977/AD977A OUTLINE DIMENSIONS Dimensions shown in inches and (mm). 20-Lead Plastic DIP (N-20) 20 11 1 10 0.280 (7.11) 0.240 (6.10) 0.325 (8.25) 0.300 (7.62) 0.195 (4.95) 0.115 (2.93) 0.060 (1.52) 0.015 (0.38) PIN 1 0.210 (5.33) MAX 0.130 (3.30) MIN 0.160 (4.06) 0.115 (2.93) 0.022 (0.558) 0.014 (0.356) 0.015 (0.381) 0.008 (0.204) 0.070 (1.77) SEATING 0.045 (1.15) PLANE 0.100 (2.54) BSC C00913d–5–10/00 (rev. D) 1.060 (26.90) 0.925 (23.50) 20-Lead Wide Body (SOIC) (R-20) 11 1 10 PIN 1 0.0118 (0.30) 0.0040 (0.10) 0.4193 (10.65) 0.3937 (10.00) 20 0.2992 (7.60) 0.2914 (7.40) 0.5118 (13.00) 0.4961 (12.60) 0.1043 (2.65) 0.0926 (2.35) 0.0291 (0.74) x 45° 0.0098 (0.25) 8° 0.0500 0.0192 (0.49) 0° (1.27) 0.0138 (0.35) SEATING 0.0125 (0.32) PLANE BSC 0.0091 (0.23) 0.0500 (1.27) 0.0157 (0.40) 28-Lead Shrink Small Outline Package (SSOP) (RS-28) 0.407 (10.34) 0.397 (10.08) 15 1 14 0.07 (1.79) 0.066 (1.67) 0.078 (1.98) PIN 1 0.068 (1.73) 0.008 (0.203) 0.0256 (0.65) 0.002 (0.050) BSC PRINTED IN U.S.A. 0.311 (7.9) 0.301 (7.64) 0.212 (5.38) 0.205 (5.21) 28 0.015 (0.38) 0.010 (0.25) SEATING 0.009 (0.229) PLANE 0.005 (0.127) –24– 8° 0° 0.03 (0.762) 0.022 (0.558) REV. D