HI5905EVAL2

HI5905N/QML TM Data Sheet July 1999 File Number 4718.1 14-Bit, 5 MSPS, Military A/D Converter The HI5905N/QML is a monolithic, 14-bit, 5MSPS Analogto-Digital Converter fabricated in an advanced BiCMOS process. It is designed for high speed, high resolution applications where wide bandwidth, low power consumption and excellent SINAD performance are essential. With a 100MHz full power input bandwidth and high frequency accuracy, the converter is ideal for many Military types of communication systems employing digital IF architectures. The HI5905N/QML is designed in a fully differential pipelined architecture with a front end differential-in-differential-out sample-and-hold amplifier (S/H). Consuming 350mW (typ) power at 5MSPS, the HI5905N/QML has excellent dynamic performance over the full Military temperature range. Data output latches are provided which present valid data to the output bus with a data latency of only 4 clock cycles. Specifications for QML devices are controlled by the Defense Supply Center in Columbus (DSCC). The SMD numbers listed below must be used when ordering. Detailed Electrical Specifications for the HI5905N/QML are contained in SMD 5962-98591. That document may be easily downloaded from our website. http://www.Intersil.com/data/sm/index.htm Features • QML Compliant per SMD 5962-9859101NXB • Sampling Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . .5MSPS • Low Power at 5MSPS. . . . . . . . . . . . . . . . . 400mW (Max) • Internal Sample and Hold • Fully Differential Architecture • Full Power Input Bandwidth . . . . . . . . . . . . . . . . . 100MHz • SINAD at 1MHz . . . . . . . . . . . . . . . . . . . . . . >69dB (Min) • Internal Voltage Reference • TTL Compatible Clock Input • CMOS Compatible Digital Data Outputs Applications • Digital Communication Systems • Undersampling Digital IF • Asymmetric Digital Subscriber Line (ADSL) • Document Scanners • Reference Literature - AN9214, Using Intersil High Speed A/D Converters - AN9785, Using the Intersil HI5905 EVAL2 Evaluation Board Pinout HI5905 (MQFP) (MO-108AA-2 ISSUE A) TOP VIEW DGND1 DVCC1 DVCC1 CLK Ordering Information ORDERING NUMBER 5962-9859101NXB HI5905EVAL2 INTERNAL INTERSIL MKT. NUMBER HI5905N/QML Low Frequency Platform TEMP. RANGE(oC) -55 to 125 25 NC NC NC NC NC DGND1 NC AVCC AGND NC NC VIN+ VINVDC 44 43 42 41 40 39 38 37 36 35 34 1 33 2 32 3 4 5 6 7 8 9 10 31 30 29 28 27 26 25 24 NC D0 D1 D2 D3 D4 D5 D6 D7 NC DVCC2 DGND2 D8 D9 NC 11 23 12 13 14 15 16 17 18 19 20 21 22 VROUT AGND AVCC VRIN D13 D12 D11 D10 NC NC 4-1 NC CAUTION: These devices are sensitive to electrostatic discharge; follow proper IC Handling Procedures. 1-888-724-7143 | Copyright © Intersil Corporation 1999 Functional Block Diagram VDC VINVIN+ S/H STAGE 1 DVCC2 BIAS CLOCK REF CLK VROUT VRIN 4-BIT FLASH + ∑ X8 - 4-BIT DAC D13 (MSB) D12 D11 DIGITAL DELAY AND DIGITAL ERROR CORRECTION D10 D9 D8 D7 D6 D5 D4 D3 D2 STAGE 4 4-BIT FLASH + ∑ X8 - 4-BIT DAC D1 D0 (LSB) STAGE 5 4-BIT FLASH DGND2 AVCC AGND DVCC1 DGND1 Typical Application Schematic (LSB) D0 (38) D1 (37) D2 (36) VRIN (14) D3 (33) AGND (6) D4 (32) AGND (15) D5 (31) DGND1 (3) D6 (30) DGND1 (42) D7 (29) DGND2 (26) D8 (25) VIN+ (9) D9 (24) D10 (21) VDC (11) D11 (20) VIN- (10) D12 (19) (MSB) D13 (18) VROUT (13) CLK (40) DVCC1 (41) AVCC (5) DVCC1 (43) +5V + 10µF 0.1µF AVCC (16)DVCC2 (27) 0.1µF + 10µF +5V D0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12 D13 DGND AGND BNC VIN+ VINCLOCK 10µF AND 0.1µF CAPS ARE PLACED AS CLOSE TO PART AS POSSIBLE HI5905 4-2 Timing Waveforms ANALOG INPUT CLOCK INPUT SN - 1 HN - 1 SN HN SN + 1 HN + 1 SN + 2 HN + 2 SN + 3 HN + 3 SN + 4 HN + 4 SN + 5 HN + 5 SN + 6 HN + 6 INPUT S/H 1ST STAGE B1 , N - 1 B1, N B1, N + 1 B1, N + 2 B1, N + 3 B1, N + 4 B1, N + 5 2ND STAGE B2 , N - 2 B2 , N - 1 B2 , N B2 , N + 1 B2 , N + 2 B2 , N + 3 B2 , N + 4 3RD STAGE 4TH STAGE B3 , N - 2 B3 , N - 1 B3 , N B3 , N + 1 B3 , N + 2 B3 , N + 3 B3 , N + 4 B4 , N - 3 B4 , N - 2 B4 , N - 1 B4 , N B4 , N + 1 B4 , N + 2 B4 , N + 3 5TH STAGE B5 , N - 3 B5 , N - 2 B5 , N - 1 B5 , N B5 , N + 1 B5 , N + 2 B5 , N + 3 DATA OUTPUT DN - 4 DN - 3 tLAT DN - 2 DN - 1 DN DN + 1 DN + 2 NOTES: 1. SN : N-th sampling period. 2. HN: N-th holding period. 3. BM, N : M-th stage digital output corresponding to N-th sampled input. 4. DN : Final data output corresponding to N-th sampled input. FIGURE 1. INTERNAL CIRCUIT TIMING ANALOG INPUT tAP tAJ CLOCK INPUT 1.5V 1.5V tOD tH 3.5V DATA N-1 1.5V DATA N DATA OUTPUT FIGURE 2. INPUT-TO-OUTPUT TIMING 4-3 Pin Descriptions PIN # 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 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 NAME NC NC DGND1 NC AVCC AGND NC NC VIN+ VINVDC NC VROUT VRIN AGND AVCC NC D13 D12 D11 D10 NC NC D9 D8 DGND2 DVCC2 NC D7 D6 D5 D4 D3 NC NC D2 D1 D0 NC CLK DVCC1 DGND1 DVCC1 NC DESCRIPTION No Connection No Connection Digital Ground No Connection Analog Supply (5.0V) Analog Ground No Connection No Connection Positive Analog Input Negative Analog Input DC Bias Voltage Output No Connection Reference Voltage Output Reference Voltage Input Analog Ground Analog Supply (5.0V) No Connection Data Bit 11 Output (MSB) Data Bit 11 Output Data Bit 11 Output Data Bit 10 Output No Connection No Connection Data Bit 9 Output Data Bit 8 Output Digital Ground Digital Supply (5.0V) No Connection Data Bit 7 Output Data Bit 6 Output Data Bit 5 Output Data Bit 4 Output Data Bit 3 Output No Connection No Connection Data Bit 2 Output Data Bit 1 Output Data Bit 0 Output (LSB) No Connection Input Clock Digital Supply (5.0V) Digital Ground Digital Supply (5.0V) No Connection Detailed Description Theory of Operation The HI5905 is a 14-bit fully differential sampling pipeline A/D converter with digital error correction. Figure 3 depicts the circuit for the front end differential-in-differential-out sampleand-hold (S/H). The switches are controlled by an internal clock which is a non-overlapping two phase signal, φ1 and φ2 , derived from the master clock. During the sampling phase, φ1 , the input signal is applied to the sampling capacitors, CS . At the same time the holding capacitors, CH , are discharged to analog ground. At the falling edge of φ1 the input signal is sampled on the bottom plates of the sampling capacitors. In the next clock phase, φ2 , the two bottom plates of the sampling capacitors are connected together and the holding capacitors are switched to the op amp output nodes. The charge then redistributes between CS and CH completing one sample-and-hold cycle. The output is a fully-differential, sampled-data representation of the analog input. The circuit not only performs the sampleand-hold function but will also convert a single-ended input to a fully-differential output for the converter core. During the sampling phase, the VIN pins see only the on-resistance of a switch and CS . The relatively small values of these components result in a typical full power input bandwidth of 100MHz for the converter. φ1 VIN + φ1 φ2 CH φ1 CS -+ +CS CH VOUT + VOUT - VIN - φ1 φ1 φ1 FIGURE 3. ANALOG INPUT SAMPLE-AND-HOLD As illustrated in the functional block diagram and the timing diagram in Figure 1, four identical pipeline subconverter stages, each containing a four-bit flash converter, a four-bit digital-to-analog converter and an amplifier with a voltage gain of 8, follow the S/H circuit with the fifth stage being only a 4-bit flash converter. Each converter stage in the pipeline will be sampling in one phase and amplifying in the other clock phase. Each individual sub-converter clock signal is offset by 180 degrees from the previous stage clock signal, with the result that alternate stages in the pipeline will perform the same operation. The output of each of the four-bit subconverter stages is a four-bit digital word containing a supplementary bit to be used by the digital error correction logic. The output of each subconverter stage is input to a digital delay line which is controlled by the internal sampling clock. The function of the digital delay line is to time align the digital outputs of the four 4-4 identical four-bit subconverter stages with the corresponding output of the fifth stage flash converter before applying the twenty bit result to the digital error correction logic. The digital error correction logic uses the supplementary bits to correct any error that may exist before generating the final fourteen bit digital data output of the converter. Because of the pipeline nature of this converter, the digital data representing an analog input sample is output to the digital data bus on the 4th cycle of the clock after the analog sample is taken. This time delay is specified as the data latency. After the data latency time, the digital data representing each succeeding analog sample is output during the following clock cycle. The digital output data is synchronized to the external sampling clock with a latch. The digital output data is available in two’s complement binary format (see Table 1, A/D Code Table). common mode voltage range of 1.0V to 4.0V. The performance of the ADC does not change significantly with the value of the analog input common mode voltage. VIN VIN+ HI5905 VDC -VIN VIN- FIGURE 4. AC COUPLED DIFFERENTIAL INPUT Internal Reference Generator, VROUT and VRIN The HI5905 has an internal reference generator, therefore, no external reference voltage is required. VROUT must be connected to VRIN when using the internal reference voltage. The HI5905 can be used with an external reference. The converter requires only one external reference voltage connected to the VRIN pin with VROUT left open. The HI5905 is tested with VROUT, equal to 4.0V, connected to VRIN . Internal to the converter, two reference voltages of 1.3V and 3.3V are generated for a fully differential input signal range of ±2V. In order to minimize overall converter noise, it is recommended that adequate high frequency decoupling be provided at the reference voltage input pin, VRIN . A 2.3V DC bias voltage source, VDC , half way between the top and bottom internal reference voltages, is made available to the user to help simplify circuit design when using a differential input. This low output impedance voltage source is not designed to be a reference but makes an excellent bias source and stays within the analog input common mode voltage range over temperature. The difference between the converter’s two internal voltage references is 2V. For the AC coupled differential input, (Figure 4), if VIN is a 2VP-P sinewave with -VIN being 180 degrees out of phase with VIN, then VIN+ is a 2VP-P sinewave riding on a DC bias voltage equal to VDC and VIN- is a 2VP-P sinewave riding on a DC bias voltage equal to VDC. Consequently, the converter will be at positive full scale, resulting in a digital data output code with D13 (MSB) equal to a logic “0” and D0-D12 equal to logic “1” (see Table 1, A/D Code Table), when the VIN+ input is at VDC+1V and the VIN- input is at VDC-1V (VIN+ - VIN- = 2V). Conversely, the ADC will be at negative full scale, resulting in a digital data output code with D13 (MSB) equal to a logic “1” and D0-D12 equal to logic “0” (see Table 1, A/D Code Table), when the VIN+ input is equal to VDC -1V and VIN- is at VDC +1V (VIN+-VIN- = -2V). From this, the converter is seen to have a peak-to-peak differential analog input voltage range of 2V . The analog input can be DC coupled (Figure 5) as long as the inputs are within the analog input common mode voltage range (1.0V ≤ VDC ≤ 4.0V). Analog Input, Differential Connection The analog input to the HI5905 can be configured in various ways depending on the signal source and the required level of performance. A fully differential connection (Figure 4) will give the best performance for the converter. Since the HI5905 is powered off a single +5V supply, the analog input must be biased so it lies within the analog input TABLE 1. A/D CODE TABLE DIFFERENTIAL CODE INPUT VOLTAGE † MSB (USING INTERNAL CENTER REFERENCE) D13 DESCRIPTION +Full Scale (+FS) - 1/4 LSB +FS - 1 1/4 LSB + 3/4 LSB - 1/4 LSB -FS + 1 3/4 LSB -Full Scale (-FS) + 3/4 LSB +1.99994V 1.99969V 183.105µV -61.035µV -1.99957V -1.99982V 0 0 0 1 1 1 TWO’S COMPLEMENT BINARY OUTPUT CODE LSB D12 1 1 0 1 0 0 D11 1 1 0 1 0 0 D10 1 1 0 1 0 0 D9 1 1 0 1 0 0 D8 1 1 0 1 0 0 D7 1 1 0 1 0 0 D6 1 1 0 1 0 0 D5 1 1 0 1 0 0 D4 1 1 0 1 0 0 D3 1 1 0 1 0 0 D2 1 1 0 1 0 0 D1 1 1 0 1 0 0 D0 1 0 0 1 1 0 † The voltages listed above represent the ideal center of each two’s complement binary output code shown. 4-5 VIN VDC R C VIN+ HI5905 VDC -VIN VDC R VIN- The resistor, R, in Figure 7 is not absolutely necessary but may be used as a load setting resistor. A capacitor, C, connected from VIN+ to VIN- will help filter any high frequency noise on the inputs, also improving performance. Values around 20pF are sufficient and can be used on AC coupled inputs as well. Note, however, that the value of capacitor C chosen must take into account the highest frequency component of the analog input signal. A single ended source will give better overall system performance if it is first converted to differential before driving the HI5905. FIGURE 5. DC COUPLED DIFFERENTIAL INPUT The resistors, R, in Figure 5 are not absolutely necessary but may be used as load setting resistors. A capacitor, C, connected from VIN+ to VIN- will help filter any high frequency noise on the inputs, also improving performance. Values around 20pF are sufficient and can be used on AC coupled inputs as well. Note, however, that the value of capacitor C chosen must take into account the highest frequency component of the analog input signal. Digital I/O and Clock Requirements The HI5905 provides a standard high-speed interface to external TTL/CMOS logic families. The digital CMOS clock input has TTL level thresholds. The low input bias current allows the HI5905 to be driven by CMOS logic. The digital CMOS outputs have a separate +5.0V digital supply input pin. In order to ensure rated performance of the HI5905, the duty cycle of the clock should be held at 50% ±5%. It must also have low jitter and operate at standard TTL levels. Performance of the HI5905 will only be guaranteed at conversion rates above 0.5MSPS. This ensures proper performance of the internal dynamic circuits. Analog Input, Single-Ended Connection The configuration shown in Figure 6 may be used with a single ended AC coupled input. Sufficient headroom must be provided such that the input voltage never goes above +5V or below AGND . VIN VIN+ Supply and Ground Considerations The HI5905 has separate analog and digital supply and ground pins to keep digital noise out of the analog signal path. The part should be mounted on a board that provides separate low impedance connections for the analog and digital supplies and grounds. For best performance, the supplies to the HI5905 should be driven by clean, linear regulated supplies. The board should also have good high frequency decoupling capacitors mounted as close as possible to the converter. If the part is powered off a single supply then the analog supply and ground pins should be isolated by ferrite beads from the digital supply and ground pins. Refer to the Application Note AN9214, “Using Intersil High Speed A/D Converters” for additional considerations when using high speed converters. VDC HI5905 VIN- FIGURE 6. AC COUPLED SINGLE ENDED INPUT Again, the difference between the two internal voltage references is 2V. If VIN is a 4VP-P sinewave, then VIN+ is a 4VP-P sinewave riding on a positive voltage equal to VDC. The converter will be at positive full scale when VIN+ is at VDC + 2V (VIN+ - VIN- = 2V) and will be at negative full scale when VIN+ is equal to VDC - 2V (VIN+ - VIN- = -2V). In this case, VDC could range between 2V and 3V without a significant change in ADC performance. The simplest way to produce VDC is to use the VDC bias voltage output of the HI5905. The single ended analog input can be DC coupled (Figure 7) as long as the input is within the analog input common mode voltage range. VIN VDC R C VDC HI5905 VIN VIN+ Static Performance Definitions Offset Error (VOS) The midscale code transition should occur at a level 1/4 LSB above half-scale. Offset is defined as the deviation of the actual code transition from this point. Full-Scale Error (FSE) The last code transition should occur for an analog input that is 3/4 LSB below positive full-scale with the offset error removed. Full-scale error is defined as the deviation of the actual code transition from this point. Differential Linearity Error (DNL) DNL is the worst case deviation of a code width from the ideal value of 1 LSB. FIGURE 7. DC COUPLED SINGLE ENDED INPUT 4-6 Integral Linearity Error (INL) INL is the worst case deviation of a code center from a best fit straight line calculated from the measured data. Power Supply Rejection Ratio (PSRR) Each of the power supplies are moved plus and minus 5% and the shift in the offset and gain error (in LSBs) is noted. present at the inputs. The ratio of the measured signal to the distortion terms is calculated. The terms included in the calculation are (f1 + f2), (f1 - f2), (2f1), (2f2), (2f1 + f2), (2f1 - f2), (f1 + 2f2), (f1 - 2f2). The ADC is tested with each tone 6dB below full scale. Transient Response Transient response is measured by providing a fullscale transition to the analog input of the ADC and measuring the number of cycles it takes for the output code to settle within 14-bit accuracy. Dynamic Performance Definitions Fast Fourier Transform (FFT) techniques are used to evaluate the dynamic performance of the HI5905. A low distortion sine wave is applied to the input, it is coherently sampled, and the output is stored in RAM. The data is then transformed into the frequency domain with an FFT and analyzed to evaluate the dynamic performance of the A/D. The sine wave input to the part is -0.5dB down from full-scale for all these tests. SNR and SINAD are quoted in dB. The distortion numbers are quoted in dBc (decibels with respect to carrier) and DO NOT include any correction factors for normalizing to full scale. Over-Voltage Recovery Over-voltage Recovery is measured by providing a fullscale transition to the analog input of the ADC which overdrives the input by 200mV, and measuring the number of cycles it takes for the output code to settle within 14-bit accuracy. Full Power Input Bandwidth (FPBW) Full power input bandwidth is the analog input frequency at which the amplitude of the digitally reconstructed output has decreased 3dB below the amplitude of the input sinewave. The input sinewave has an amplitude which swings from -fS to +fS . The bandwidth given is measured at the specified sampling frequency. Signal-to-Noise Ratio (SNR) SNR is the measured RMS signal to RMS noise at a specified input and sampling frequency. The noise is the RMS sum of all of the spectral components except the fundamental and the first five harmonics. Signal-to-Noise + Distortion Ratio (SINAD) SINAD is the measured RMS signal to RMS sum of all other spectral components below the Nyquist frequency, fS/2, excluding DC. Timing Definitions Refer to Figure 1, Internal Circuit Timing, and Figure 2, Input-To-Output Timing, for these definitions. Aperture Delay (tAP) Aperture delay is the time delay between the external sample command (the falling edge of the clock) and the time at which the signal is actually sampled. This delay is due to internal clock path propagation delays. Effective Number Of Bits (ENOB) The effective number of bits (ENOB) is calculated from the SINAD data by: ENOB = ( SINAD + V CORR -1.76 ) /6.02 where: VCORR = 0.5dB (Typical) VCORR adjusts the ENOB for the amount the input is below fullscale. Aperture Jitter (tAJ) Aperture Jitter is the RMS variation in the aperture delay due to variation of internal clock path delays. Total Harmonic Distortion (THD) THD is the ratio of the RMS sum of the first 5 harmonic components to the RMS value of the fundamental input signal. Data Hold Time (tH) Data hold time is the time to where the previous data (N - 1) is still valid. 2nd and 3rd Harmonic Distortion This is the ratio of the RMS value of the applicable harmonic component to the RMS value of the fundamental input signal. Data Output Delay Time (tOD) Data output delay time is the time to where the new data (N) is valid. Spurious Free Dynamic Range (SFDR) SFDR is the ratio of the fundamental RMS amplitude to the RMS amplitude of the next largest spur or spectral component (excluding the first 5 harmonic components) in the spectrum below fS/2. Data Latency (tLAT) After the analog sample is taken, the digital data is output on the bus at the third cycle of the clock. This is due to the pipeline nature of the converter where the data has to ripple through the stages. This delay is specified as the data latency. After the data latency time, the data representing each succeeding sample is output at the following clock pulse. The digital data lags the analog input sample by 4 clock cycles. Intermodulation Distortion (IMD) Nonlinearities in the signal path will tend to generate intermodulation products when two tones, f1 and f2 , are 4-7 HI5905N/QML Metric Plastic Quad Flatpack Packages (MQFP/PQFP) D D1 -D- Q44.10x10 (JEDEC MO-108AA-2 ISSUE A) 44 LEAD METRIC PLASTIC QUAD FLATPACK PACKAGE SYMBOL A A1 A2 INCHES MIN 0.004 0.077 0.012 0.012 0.510 0.390 0.510 0.390 0.026 44 0.032 BSC MAX 0.093 0.010 0.083 0.018 0.016 0.530 0.398 0.530 0.398 0.037 MILLIMETERS MIN 0.10 1.95 0.30 0.30 12.95 9.90 12.95 9.90 0.65 44 0.80 BSC MAX 2.35 0.25 2.10 0.45 0.40 13.45 10.10 13.45 10.10 0.95 NOTES 6 3 4, 5 3 4, 5 7 Rev. 1 1/94 NOTES: 1. Controlling dimension: MILLIMETER. Converted inch dimensions are not necessarily exact. 2. All dimensions and tolerances per ANSI Y14.5M-1982. 3. Dimensions D and E to be determined at seating plane -C- . 4. Dimensions D1 and E1 to be determined at datum plane -H- . 5. Dimensions D1 and E1 do not include mold protrusion. Allowable protrusion is 0.25mm (0.010 inch) per side. 6. Dimension B does not include dambar protrusion. Allowable dambar protrusion shall be 0.08mm (0.003 inch) total. 7. “N” is the number of terminal positions. 0.13/0.23 0.005/0.009 -AE E1 -B- B B1 D D1 E e PIN 1 SEATING A PLANE 0.10 0.004 5o-16o 0.40 0.016 MIN 0o MIN 0o-7o A2 A1 0.20 M C A-B S 0.008 -CDS B B1 0.13/0.17 0.005/0.007 BASE METAL WITH PLATING E1 L N e -H- L 5o-16o All Intersil semiconductor products are manufactured, assembled and tested under ISO9000 quality systems certification. Intersil semiconductor products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design and/or specifications at any time without notice. Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Intersil is believed to be accurate and reliable. However, no responsibility is assumed by Intersil or its subsidiaries 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 Intersil or its subsidiaries. For information regarding Intersil Corporation and its products, see web site www.intersil.com Sales Office Headquarters NORTH AMERICA Intersil Corporation P. O. Box 883, Mail Stop 53-204 Melbourne, FL 32902 TEL: (321) 724-7000 FAX: (321) 724-7240 EUROPE Intersil SA Mercure Center 100, Rue de la Fusee 1130 Brussels, Belgium TEL: (32) 2.724.2111 FAX: (32) 2.724.22.05 ASIA Intersil Ltd. 8F-2, 96, Sec. 1, Chien-kuo North, Taipei, Taiwan 104 Republic of China TEL: 886-2-2515-8508 FAX: 886-2-2515-8369 4-8