View EVAL-AD7864-3CB3_6532745.PDF datasheet online --- IC-ON-LINE

Datasheet File OCR Text:

4-channel, simultaneous sampling, high speed, 12-bit adc ad7864 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 that may result from its use. specifications subject to change without notice. no license is granted by implication or otherwise under any patent or patent rights of analog devices. trademarks and registered trademarks are the property of their respective owners. one technology way, p.o. box 9106, norwood, ma 02062-9106, u.s.a. tel: 781.329.4700 www.analog.com fax: 781.461.3113 ?1998C2009 analog devices, inc. all rights reserved. features high speed (1.65 s) 12-bit adc 4 simultaneously sampled inputs 4 track-and-hold amplifiers 0.35 s track-and-hold acquisition time 1.65 s conversion time per channel hw/sw select of channel sequence for conversion single-supply operation selection of input ranges 10 v, 5 v for ad7864-1 2.5 v for ad7864-3 0 v to 2.5 v, 0 v to 5 v for ad7864-2 high speed parallel interface that allows interfacing to 3 v processors low power, 90 mw typical power saving mode, 20 w typical overvoltage protection on analog inputs applications ac motor control uninterrupted power supplies data acquisition systems communications general description the ad7864 is a high speed, low power, 4-channel, simulta- neous sampling 12-bit analog-to-digital converter (adc) that operates from a single 5 v supply. the part contains a 1.65 s successive approximation adc, four track-and-hold amplifiers, a 2.5 v reference, an on-chip clock oscillator, signal conditioning circuitry, and a high speed parallel interface. the input signals on four channels sample simultaneously preserving the relative phase information of the signals on the four analog inputs. the part accepts analog input ranges of 10 v, 5 v (ad7864-1), 0 v to +2.5 v, 0 v to +5 v (ad7864-2), and 2.5 v (ad7864-3). any subset of the four channels can be converted to maximize the throughput rate on the selected sequence. select the channels to convert via hardware (channel select input pins) or software (pro- gramming the channel select register). a single conversion start signal ( convst ) simultaneously places all the track-and-holds into hold and initiates a conversion se- quence for the selected channels. the eoc signal indicates the end of each individual conversion in the selected conversion sequence. the busy signal indicates the end of the conversion sequence. functional block diagram stby frstdata int/ext clock select output data registers 2.5v reference signal scaling v in1a v in1b v in2a v in2b v in3a v in3b v in4a v in4b busy signal scaling signal scaling signal scaling track-and-hold 4 eoc wr cs db0 db11 rd agnd dgnd v drive dv dd v ref gnd v ref a v dd 6k ? ad7864 12-bit adc software latch conversion control logic mux convst sl2 sl1 sl3 sl4 h/s sel clkin int/ext clk agnd agnd int clock db0 to db3 01341-001 figure 1. data is read from the part by a 12-bit parallel data bus using the standard cs and rd signals. maximum throughput for a single channel is 500 ksps. for all four channels, the maximum throughput is 130 ksps for the read-during-conversion sequence operation. the throughput rate for the read-after-conversion sequence operation depends on the read cycle time of the processor. see the section. the ad7864 is available in a small (0.3 square inch area) 44-lead mqfp. timing and control product highlights 1. four track-and-hold amplifiers and a fast (1.65 s) adc for simultaneous sampling and conversion of any subset of the four channels. 2. a single 5 v supply consuming only 90 mw typical, makes it ideal for low power and portable applications. see the standby mode operation section. 3. high speed parallel interface for easy connection to micro- processors, microcontrollers, and digital signal processors. 4. available in three versions with different analog input ranges. the ad7864-1 offers the standard industrial input ranges of 10 v and 5 v; the ad7864-3 offers the common signal processing input range of 2.5 v; the ad7864-2 can be used in unipolar, 0 v to 2.5 v and 0 v to 5 v, applications. 5. features very tight aperture delay matching between the four input sample-and-hold amplifiers.
ad7864 rev. d | page 2 of 28 table of contents features .............................................................................................. 1 applications ....................................................................................... 1 general description ......................................................................... 1 functional block diagram .............................................................. 1 product highlights ........................................................................... 1 revision history ............................................................................... 2 specifications ..................................................................................... 3 timing characteristics ................................................................ 5 absolute maximum ratings ............................................................ 6 esd caution .................................................................................. 6 pin configuration and function descriptions ............................. 7 terminology ...................................................................................... 9 theory of operation ...................................................................... 11 converter details ........................................................................ 11 circuit description ......................................................................... 13 analog input ............................................................................... 13 selecting a conversion sequence ................................................. 15 timing and control ................................................................... 15 using an external clock ............................................................ 17 standby mode operation .......................................................... 18 accessing the output data registers ....................................... 18 offset and full-scale adjustment ................................................ 20 positive full-scale adjust .......................................................... 20 negative full-scale adjust ......................................................... 20 dynamic specifications ................................................................. 21 signal-to-noise ratio (snr)..................................................... 21 effective number of bits ........................................................... 21 intermodulation distortion ...................................................... 21 ac linearity plots ...................................................................... 22 measuring aperture jitter .......................................................... 22 microprocessor interfacing ........................................................... 24 ad7864 to adsp-2100/adsp-2101/adsp-2102 interface . 24 ad7864 to tms320c5x interface ............................................ 24 ad7864 to mc68hc000 interface .......................................... 24 vector motor control ................................................................ 25 multiple ad7864s in a system ................................................. 26 outline dimensions ....................................................................... 27 ordering guide .......................................................................... 27 revision history 2/09rev. c to rev. d change to t 2 parameter, table 2 ...................................................... 5 2/09rev. b to rev. c updated format .................................................................. universal changes to t 5 timing parameter, table 2....................................... 5 changes to figure 15 ...................................................................... 20 changes to ad7864 to mc68hc000 interface section ............ 24 changes to figure 25 ...................................................................... 24 updated outline dimensions ....................................................... 29 changes to ordering guide .......................................................... 29 3/04rev. a to rev. b. changes to specifications and to footnote 4 ................................. 2 changes to timing characteristics footnote 1 ............................. 4 addition to absolute maximum ratings ....................................... 5 changes to ordering guide ............................................................. 5 changes to figure 7 ......................................................................... 11 changes to figure 11 ...................................................................... 13 updated outline dimensions ....................................................... 19 added revision history ................................................................ 20 updated publication code ............................................................ 20
ad7864 rev. d | page 3 of 28 specifications v dd = 5 v 5%, agnd = dgnd = 0 v, v ref = internal, clock = internal; all specifications t min to t max , unless otherwise noted. table 1. parameter a version 1 b version unit test conditions/comments sample and hold ?3 db full power bandwidth 3 3 mhz typ aperture delay 20 20 ns max aperture jitter 50 50 ps max aperture delay matching 4 4 ns max dynamic performance 2 f in = 100.0 khz, f s = 500 ksps signal-to-(noise + distortion) ratio 3 @ 25c 70 72 db min t min to t max 70 70 db min total harmonic distortion 3 ?80 ?80 db max peak harmonic or spurious noise 3 ?80 ?80 db max intermodulation distortion 3 fa = 49 khz, fb = 50 khz second-order terms ?80 ?80 db typ third-order terms ?80 ?80 db typ channel-to-channel isolation 3 ?80 ?80 db max f in = 50 khz sine wave dc accuracy any channel resolution 12 12 bits relative accuracy 3 1 1/2 lsb max differential nonlinearity 3 0.9 0.9 lsb max no missing codes ad7864-1 positive gain error 3 3 3 lsb max positive gain error match 3 +3 3 lsb max negative gain error 3 3 3 lsb max negative gain error match 3 +3 3 lsb max bipolar zero error 4 3 lsb max bipolar zero error match +2 2 lsb max ad7864-3 positive gain error 3 3 lsb max positive gain error match 3 2 lsb max negative gain error 3 3 lsb max negative gain error match 3 2 lsb max bipolar zero error 3 lsb max bipolar zero error match 2 lsb max ad7864-2 positive gain error 3 3 lsb max positive gain error match 3 3 lsb max unipolar offset error 3 lsb max unipolar offset error match 2 lsb max analog inputs ad7864-1 input voltage range 5, 10 5, 10 v input resistance 9, 18 9, 18 k min ad7864-3 input voltage range 2.5 2.5 v input resistance 4.5 4.5 k min
ad7864 rev. d | page 4 of 28 parameter a version 1 b version unit test conditions/comments ad7864-2 input voltage range 0 to 2.5, 0 to 5 0 to 2.5, 0 to 5 v input current (0 v to 2.5 v option) 100 100 na max input resistance (0 v to 5 v option) 9 9 k min reference input/output v ref in input voltage range 2.375/2.625 2.375/2.625 v min /v max 2.5 v 5% v ref in input capacitance 4 10 10 pf max v ref out output voltage 2.5 2.5 v nom v ref out error @ 25c 10 10 mv max v ref out error t min to t max 20 20 mv max v ref out temperature coefficient 25 25 ppm/c typ v ref out output impedance 6 6 k typ see the reference section logic inputs input high voltage, v inh 2.4 2.4 v min v dd = 5 v 5% input low voltage, v inl 0.8 0.8 v max v dd = 5 v 5% input current, i in 10 10 a max input capacitance, c in 4 10 10 pf max logic outputs output high voltage, v oh 4.0 4.0 v min i source = 400 a output low voltage, v ol 0.4 0.4 v max i sink = 1.6 ma db11 to db0 high impedance leakage current 10 10 a max capacitance 4 10 10 pf max output coding ad7864-1, ad7864-3 twos complement ad7864-2 straight (natural) binary conversion rate conversion time 1.65 1.65 s max for one channel track-and-hold acquisition time 2 , 3 0.35 0.35 s max throughput time 130 130 ksps max for all four channels power requirements v dd 5 5 v nom 5% for specified performance i dd 5 a typical, logic inputs = 0 v or v dd normal mode 24 24 ma max standby mode 20 20 a max typically 4 a power dissipation normal mode 120 120 mw max typically 90 mw standby mode 100 100 w max typically 20 w 1 temperature ranges are as follows: a, b versions: C40c to +85c. the a version is fully specif ied up to 105c with a maximum sample rate of 450 ksps and i dd maximum (normal mode) of 26 ma. 2 performance is measured through the full channel (sha and adc). 3 see the terminology section. 4 sample tested at initial release to ensure compliance.
ad7864 rev. d | page 5 of 28 timing characteristics v drive = 5 v 5%, agnd = dgnd = 0 v, v ref = internal, clock = internal; all specifications t min to t max , unless otherwise noted. 1 , 2 table 2. parameter a, b versions unit test conditions/comments t conv 1.65 s max conversion time, internal clock 13 clock cycles conversion time, external clock 2.6 s max clkin = 5 mhz t acq 0.34 s max acquisition time t busy no. of channels (t conv + t 9 ) ? t 9 s max selected number of channels multiplied by (t conv + eoc pulse width) eoc pulse width t wake-up external v ref 2 s max stby rising edge to convst rising edge t wake-up internal v ref 3 6 ms max stby rising edge to convst rising edge t 1 35 ns min convst pulse width t 2 70 ns max convst rising edge to busy rising edge read operation t 3 0 ns min cs to rd setup time t 4 0 ns min cs to rd hold time t 5 35 ns min read pulse width, v drive = 5 v 40 ns min read pulse width, v drive = 3 v t 6 4 35 ns max data access time after falling edge of rd , v drive = 5 v 40 ns max data access time after falling edge of rd , v drive = 3 v t 7 5 5 ns min bus relinquish time after rising edge of rd 30 ns max t 8 10 ns min time between consecutive reads t 9 75 ns min eoc pulse width 180 ns max t 10 70 ns max rd rising edge to frstdata edge (rising or falling) t 11 15 ns max eoc falling edge to frstdata falling delay t 12 0 ns min eoc to rd delay write operation t 13 20 ns min wr pulse width t 14 0 ns min cs to wr setup time t 15 0 ns min wr to cs hold time t 16 5 ns min input data setup time of rising edge of wr t 17 5 ns min input data hold time 1 sample tested at initial release to ensure compliance. all input signals are measured with tr = tf = 1 ns (10% to 90% of 5 v) and timed from a voltage level of 1.6 v. 2 see figure 9, figure 10,and figure 11. 3 refer to the standby mode operation sectio n. the maximum specification of 6 ms is vali d when using a 0.1 f decoupling capacit or on the v ref pin. 4 measured with the load circuit of figure 2 and defined as the time required for an output to cross 0.8 v or 2.4 v. 5 these times are derived from the measured time taken by the data outputs to change 0.5 v when loaded with the circuit shown in figure 2. the measured number is then extrapolated back to remove the effects of charging or discharging the 50 pf capacitor. this means that the times quoted i n the timing characteristics are the true bus relinquish times of the part, and as such, ar e independent of external bus loading capacitances. to o utput 50pf 1.6v 400a 1.6ma 01341-002 figure 2. load circuit for access time and bus relinquish time
ad7864 rev. d | page 6 of 28 absolute maximum ratings t a = 25c, unless otherwise noted. table 3. parameter rating av dd to agnd ?0.3 v to +7 v dv dd to dgnd ?0.3 v to +7 v agnd to dgnd ?0.3 v to +0.3 v av dd to dv dd ?0.3 v to +0.3 v analog input voltage to agnd ad7864-1 (10 v input range) 20 v ad7864-1 (5 v input range) ?7 v to +20 v ad7864-3 ?7 v to +20 v ad7864-2 ?1 v to +20 v reference input voltage to agnd ?0.3 v to v dd + 0.3 v digital input voltage to dgnd ?0.3 v to v dd + 0.3 v digital output voltage to dgnd ?0.3 v to v dd + 0.3 v v drive to agnd ?0.3 v to av dd + 0.3 v v drive to dgnd ?0.3 v to dv dd + 0.3 v operating temperature range commercial (a and b versions) ?40c to +85c storage temperature range ?65c to +150c junction temperature 150c mqfp package, power dissipation 450 mw ja thermal impedance 95c/w lead temperature, soldering vapor phase (60 sec) 215c infrared (15 sec) 220c 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. esd caution
ad7864 rev. d | page 7 of 28 db7 db8 db9 db10 db11 clkin int/ext clk busy frstdat pin configuration and fu nction descriptions a convst cs rd wr sl1 sl2 sl3 sl4 h/s sel agnd av dd v ref v ref gnd v in2a v in1b v in1a stby agnd v in4b v in2b agnd v in4a eoc db0 db1 db3 db4 db5 dgnd v drive dv dd db2 db6 v in3b v in3a pin 1 ad7864 top view (not to scale) 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 3536 37 38 39 4041424344 01341-003 figure 3. pin configuration table 4. pin function descriptions pin no. mnemonic description 1 busy busy output. the busy output is tri ggered high by the rising edge of convst and remains high until conversion is completed on all selected channels. 2 frstdata first data output. frstdata is a logic output which, when high, indica tes that the output data register pointer is addressing register 1see the accessing the output data registers section. 3 convst convert start input. logic input. a low-to-high transition on this input puts all trac k-and-holds into their hold mode and starts conversion on the selected channels. in addition, the state of the channel sequence selection is also latched on the rising edge of convst . 4 cs chip select input. active low logic input. the d evice is selected when this input is active. 5 rd read input. active low logic input that is used in conjunction with cs low to enable the data outputs. ensure the wr pin is at logic high while performing a read operation. 6 wr write input. a rising edge on the wr input, with cs low and rd high, latches the logic state on db0 to db3 into the channel select register. 7 to 10 sl1 to sl4 hardware channel select. conversion sequence selection can also be made via the sl1 to sl4 pins if h /s sel is logic 0. the selection is latched on the rising edge of convst . see the section. selecting a conversion sequence 11 h /s sel hardware/software select input. when this pin is at logic 0, the ad7864 conversion sequence selection is controlled via the sl1 to sl4 input pins. when this pin is at logic 1, the sequence is controlled via the channel select register. see the selecting a conversion sequence section. 12 agnd analog ground. general analog ground. connect this agnd pin to the agnd plane of the system. 13 to 16 v in4x , v in3x analog inputs. see the analog input section. 17 agnd analog ground. analog ground reference for the atte nuator circuitry. connect this agnd pin to the agnd plane of the system. 18 to 21 v in2x , v in1x analog inputs. see the analog input section. 22 stby standby mode input. ttl-compatible in put that is used to put the device into the power save or standby mode. the stby input is high for normal operati on and low for standby operation. 23 v ref gnd reference ground. this is the ground reference for the on-chip reference buffer of the part. connect the v ref gnd pin to the agnd plane of the system. 24 v ref reference input/output. this pin provides access to th e internal reference (2.5 v 5%) and also allows the internal reference to be overdriven by an external reference source (2.5 v). connect a 0.1 f decoupling capacitor between this pin and agnd. 25 av dd analog positive supply voltage, 5.0 v 5%. 26 agnd analog ground. analog ground reference for the dac circuitry.
ad7864 rev. d | page 8 of 28 pin no. mnemonic description 27 int /ext clk internal/external clock select input. when this pin is at logic 0, the ad7864 uses its internally generated master clock. when this pin is at logic 1, the master clock is generated externally to the device. 28 clkin conversion clock input. this is an externally applied clock that allows the user to control the conversion rate of the ad7864. each conversion needs 14 clock cycles for the conversion to be completed and an eoc pulse to be generated. the clock should have a duty cycle that is no worse than 60/40. see the section. using an external clock 29 to 34 db11 to db6 data bit 11 is the msb, followed by da ta bit 10 to data bit 6. three-state ttl outputs. output coding is twos complement for the ad7864-1 and ad7864-3. output coding is straight (natural) binary for the ad7864-2. 35 dv dd positive supply voltage for digital section, 5.0 v 5% . connect a 0.1 f decoupling capacitor between this pin and agnd. both dv dd and av dd should be externally tied together. 36 v drive this pin provides the positive supply voltage for the output drivers (db0 to db11), busy, eoc , and frstdata. it is normally tied to dv dd . decouple v drive with a 0.1 f capacitor to improve performance when reading during the conversion sequence. to facilitate interfacing to 3 v processors and dsps, the outp ut data drivers can also be powered by a 3 v 10% supply. 37 dgnd digital ground. this is the ground reference for digital circuitry. connect this dgnd pin to the agnd plane of the system at the agnd pin. 38, 39 db5, db4 data bit 5 to data bit 4. three-state ttl outputs. 40 to 43 db3 to db0 data bit 3 to data bit 0. bidirectional data pins. when a read operation takes place, these pins are thre e-state ttl outputs. the channel select register is programmed wi th the data on the db0 to db3 pins with standard cs and wr signals. db0 represents channel 1, and db3 represents channel 4. 44 eoc end-of-conversion. active low logic output indicating conversion status. the end of each conversion in a conversion sequence is indicated by a low-going pulse on this line.
ad7864 rev. d | page 9 of 28 terminology channel-to-channel isolation signal-to-(noise + distortion) ratio channel-to-channel isolation is a measure of the level of crosstalk between channels. it is measured by applying a full- scale 50 khz sine wave signal to all nonselected input channels and determining how much that signal is attenuated in the selected channel. the figure given is the worst case across all four channels. this is the measured ratio of signal-to-(noise + distortion) at the output of the adc. the signal is the rms amplitude of the fundamental. noise is the rms sum of all nonfundamental signals up to half the sampling frequency (f s /2), excluding dc. the ratio depends on the number of quantization levels in the digitization process; the more levels, the smaller the quantization noise. the theoretical signal-to-(noise + distortion) ratio for an ideal n-bit converter with a sine wave input is given by relative accuracy relative accuracy, or endpoint nonlinearity, is the maximum deviation from a straight line passing through the endpoints of the adc transfer function. signal-to-(noise + distortion) = (6.02 n + 1.76) db thus, for a 12-bit converter, this is 74 db. differential nonlinearity total harmonic distortion (thd) this is the difference between the measured and the ideal 1 lsb change between any two adjacent codes in the adc. thd is the ratio of the rms sum of harmonics to the fundamental. for the ad7864, it is defined as positive full-scale error 1 6 54 32 v vvvvv thd 22222 log20)db( ++++ = this is the deviation of the last code transition (01...110 to 01...111) from the ideal, 4 v ref ? 3/2 lsb (ad7864-1, 10 v), or 2 v ref ? 3/2 lsb (ad7864-1, 5 v range), or v ref ? 3/2 lsb (ad7864-3, 2.5 v range), after the bipolar offset error has been adjusted out. where v 1 is the rms amplitude of the fundamental, and v 2 , v 3 , v 4 , v 5 , and v 6 are the rms amplitudes of the second through the fifth harmonics. positive full-scale error (ad7864-2, 0 v to 2.5 v and 0 v to 5 v) peak harmonic or spurious noise this is the deviation of the last code transition (11...110 to 11...111) from the ideal 2 v ref ? 3/2 lsb (ad7864-2, 0 v to 5 v range) or v ref ? 3/2 lsb (ad7864-2, 0 v to 2.5 v range), after the unipolar offset error has been adjusted out. peak harmonic or spurious noise is defined as the ratio of the rms value of the next largest component in the adc output spectrum (up to f s /2 and excluding dc) to the rms value of the fundamental. normally, the value of this specification is deter- mined by the largest harmonic in the spectrum, but for parts where the harmonics are buried in the noise floor, it is a noise peak. bipolar zero error (ad7864-1, 10 v/5 v, ad7864-3, 2.5 v) this is the deviation of the midscale transition (all 0s to all 1s) from the ideal, agnd ? 1/2 lsb. intermodulation distortion unipolar offset error (ad7864-2, 0 v to 2.5 v and 0 v to 5 v) with inputs consisting of sine waves at two frequencies, fa and fb, any active device with nonlinearities creates distortion products at sum and difference frequencies of mfa nfb, where m, n = 0, 1, 2, 3, and so on. intermodulation terms are those for which neither m nor n are equal to zero. for example, second-order terms include (fa + fb) and (fa ? fb), whereas third-order terms include (2 fa + fb), (2 fa ? fb), (fa + 2 fb), and (fa ? 2 fb). this is the deviation of the first code transition (00...000 to 00...001) from the ideal, agnd + 1/2 lsb. negative full-scale error (ad7864-1, 10 v/5 v, and ad7864-3, 2.5 v) this is the deviation of the first code transition (10...000 to 10...001) from the ideal, ?4 v ref + 1/2 lsb (ad7864-1, 10 v), ?2 v ref + 1/2 lsb (ad7864-1, 5 v range) or ?v ref + 1/2 lsb (ad7864-3, 2.5 v range), after bipolar zero error has been adjusted out. the ad7864 is tested using the ccif standard, where two input frequencies near the top end of the input bandwidth are used. in this case, the second- and third-order terms are of different significance. the second-order terms are usually distanced in frequency from the original sine waves, whereas the third-order terms are usually at a frequency close to the input frequencies. as a result, the second- and third-order terms are specified separately. the calculation of the intermodulation distortion is as per the thd specification where it is the ratio of the rms sum of the individual distortion products to the rms amplitude of the funda- mental expressed in decibels. track-and-hold acquisition time track-and-hold acquisition time is the time required for the output of the track-and-hold amplifier to reach its final value, within 1/2 lsb, after the end of a conversion (the point at which the track-and-hold returns to track mode). it also applies to situations where there is a step input change on the input voltage applied to the selected v inxa /v inxb input of the ad7864.
ad7864 rev. d | page 10 of 28 it means that the user must wait for the duration of the track- and-hold acquisition time after the end of conversion or after a step input change to v inxa /v inxb before starting another conversion to ensure that the part operates to specification.
ad7864 rev. d | page 11 of 28 theory of operation converter details the ad7864 is a high speed, low power, 4-channel simultaneous sampling 12-bit adc that operates from a single 5 v supply. the part contains a 1.65 s successive approximation adc, four track-and-hold amplifiers, an internal 2.5 v reference, and a high speed parallel interface. there are four analog inputs that can be simultaneously sampled, thus preserving the relative phase information of the signals on all four analog inputs. thereafter, conversions are completed on the selected subset of the four channels. the part accepts an analog input range of 10 v or 5 v (ad7864-1), 2.5 v (ad7864-3), and 0 v to +2.5 v or 0 v to +5 v (ad7864-2). overvoltage protection on the analog inputs of the part allows the input voltage to go to 20 v, (ad7864-1 10 v range), ?7 v or +20 v (ad7864-1 5 v range), ?1 v to +20 v (ad7864-2), and ?7 v to +20 v (ad7864-3), without causing damage. the ad7864 has two operating modes: reading-between-conversions and reading- after-the-conversion sequence. these modes are discussed in more detail in the timing a n d control section. a conversion is initiated on the ad7864 by pulsing the convst input. on the rising edge of convst , all four on-chip track- and-holds are placed into hold simultaneously and the conversion sequence is started on all the selected channels. channel selection is made via the sl1 to sl4 pins if h /s sel is logic 0 or via the channel select register if h /s sel is logic 1see the section. the channel select register is programmed via the bidirectional data lines (db0 to db3) and a standard write operation. the selected conversion sequence is latched on the rising edge of selecting a conversion sequence convst , therefore, changing a selection only takes effect once a new conversion sequence is initiated. the busy output signal is triggered high on the rising edge of convst and remains high for the duration of the conver- sion sequence. the conversion clock for the part is generated internally using a laser trimmed, clock oscillator circuit. there is also the option of using an external clock, by tying the int /ext clk pin logic high, and applying an external clock to the clkin pin. however, the optimum throughput is obtained by using the internally generated clocksee the section. the using an external clock eoc signal indicates the end of each conversion in the conversion sequence. the busy signal indicates the end of the full conversion sequence, and at this time, all four track and holds return to tracking mode. the conversion results can be read either at the end of the full conversion sequence (indicated by busy going low), or as each result becomes available (indicated by eoc going low). data is read from the part via a 12-bit parallel data bus with standard cs and rd signalssee the section. timing and control conversion time for each channel of the ad7864 is 1.65 s, and the track-and-hold acquisition time is 0.35 s. to obtain optimum performance from the part, the read operation should not occur during a channel conversion or during the 100 ns prior to the next convst rising edge. this allows the part to operate at throughput rates up to 130 khz for all four channels and achieve data sheet specifications. track-and-hold amplifiers the track-and-hold amplifiers on the ad7864 allow the adcs to accurately convert an input sine wave of full-scale amplitude to 12-bit accuracy. the input bandwidth of the track-and-hold is greater than the nyquist rate of the adc even when the adc is operated at its maximum throughput rate of 500 ksps (that is, the track-and-hold can handle input frequencies in excess of 250 khz). the track-and-hold amplifiers acquire input signals to 12-bit accuracy in less than 350 ns. the operation of the track-and- holds are essentially transparent to the user. the four track-and- hold amplifiers sample their respective input channels simulta- neously, on the rising edge of convst . the aperture time for the track-and-holds (that is, the delay time between the external convst signal and the track-and-hold actually going into hold) is typically 15 ns and, more importantly, is well matched across the four track-and-holds on one device as well as being well matched from device to device. this allows the relative phase information between different input channels to be accurately preserved. it also allows multiple ad7864s to sample more than four channels simultaneously. at the end of a conversion sequence, the part returns to its tracking mode. the acquisition time of the track-and-hold amplifiers begin at this point. reference the ad7864 contains a single reference pin, labeled v ref . the v ref pin provides access to the 2.5 v reference within the part, or it serves as the reference source for the part by connecting v ref to an external 2.5 v reference. the part is specified with a 2.5 v reference voltage. errors in the reference source result in gain errors in the transfer function of the ad7864 and adds to the specified full-scale errors on the part. on the ad7864-1 and ad7864-3, it also results in an offset error injected in the attenuator stage; see figure 4 and figure 6 . the ad7864 contains an on-chip 2.5 v reference. to use this reference as the reference source for the ad7864, simply con- nect a 0.1 f disk ceramic capacitor from the v ref pin to agnd. the voltage that appears at this pin is internally buffered before being applied to the adc. if this reference is used externally to the ad7864, it should be buffered because the part has a fet switch in series with the reference output resulting in a 6 k
ad7864 rev. d | page 12 of 28 nominal source impedance for this output. the tolerance on the internal reference is 10 mv at 25c with a typical temperature coefficient of 25 ppm/c and a maximum error overtemperature of 20 mv. if the application requires a reference with a tighter tolerance or the ad7864 needs to be used with a system reference, the user has the option of connecting an external reference to this v ref pin. the external reference effectively overdrives the internal reference and thus provides the reference source for the adc. the reference input is buffered before being applied to the adc with the maximum input current of 100 a. suitable reference sources for the ad7864 include the ad680, ad78 0, ref192, and ref43 precision 2.5 v references.
ad7864 rev. d | page 13 of 28 circuit description analog input table 5. ideal input/output code table for the ad7864-1 analog input 1 digital output code transition +fsr/2 ? 3/2 lsb 2 011...110 to 011...111 +fsr/2 ? 5/2 lsb 011...101 to 011...110 +fsr/2 ? 7/2 lsb 011...100 to 011...101 agnd + 3/2 lsb 000...001 to 000...010 agnd + 1/2 lsb 000...000 to 000...001 agnd ? 1/2 lsb 111...111 to 000...000 agnd ? 3/2 lsb 111...110 to 111...111 ?fsr/2 + 5/2 lsb 100...010 to 100...011 ?fsr/2 + 3/2 lsb 100...001 to 100...010 ?fsr/2 + 1/2 lsb 100...000 to 100...001 the ad7864 is offered in three models: the ad7864-1, where each input can be configured for 10 v or a 5 v input voltage range; the ad7864-3, which handles the input voltage range of 2.5 v; and the ad7864-2, where each input can be configured to have a 0 v to +2.5 v or 0 v to +5 v input voltage range. ad7864-1 figure 4 shows the analog input section of the ad7864-1. each input can be configured for 5 v or 10 v operation on the ad7864-1. for 5 v (ad7864-1) operation, the v inxa and v inxb inputs are tied together and the input voltage is applied to both. for 10 v (ad7864-1) operation, the v inxb input is tied to agnd and the input voltage is applied to the v inxa input. the v inxa and v inxb inputs are symmetrical and fully interchangeable. thus for ease of printed circuit board (pcb) layout on the 10 v range, the input voltage may be applied to the v inxb input while the v inxa input is tied to agnd. 1 fsr is full-scale range and is 20 v for the 10 v range and +10 v for the 5 v range, with v ref = 2.5 v. 2 1 lsb = fsr/4096 = 4.883 mv (10 v fo r the ad7864-1) and 2.441 mv (5 v for the ad7864-1) with v ref = 2.5 v. ad7864-2 figure 5 shows the analog input section of the ad7864-2. each input can be configured for 0 v to 5 v operation or 0 v to 2.5 v operation. for 0 v to 5 v operation, the v inxb input is tied to agnd and the input voltage is applied to the v inxa input. for 0 v to 2.5 v operation, the v inxa and v inxb inputs are tied together and the input voltage is applied to both. the v inxa and v inxb inputs are symmetrical and fully interchangeable. thus for ease of pcb layout on the 0 v to 5 v range, the input voltage may be applied to the v inxb input while the v inxa input is tied to agnd. 2.5v reference t/h to adc reference circuitry 6k? r2 r3 to internal comparator ad7864-1 r1 r4 agnd v in1b v in1a v ref 01341-004 for the ad7864-2, r1 = 6 k and r2 = 6 k. the designed code transitions occur on successive integer least significant bit values. output coding is straight (natural) binary with 1 lsb = fsr/4096 = 2.5 v/4096 = 0.61 mv, and 5 v/4096 = 1.22 mv, for the 0 v to 2.5 v and 0 v to 5 v options, respectively. figure 4. ad7864-1 analog input structure table 6 shows the ideal input and output transfer function for the ad7864-2. for the ad7864-1, r1 = 6 k, r2 = 24 k, r3 = 24 k, and r4 = 12 k. the resistor input stage is followed by the high input impedance stage of the track-and-hold amplifier. 2.5v reference t/h to adc reference circuitry r2 to internal comparator ad7864-2 r1 6k? v in1b v in1a v ref 01341-005 the designed code transitions take place midway between successive integer least significant bit values (that is, 1/2 lsb, 3/2 lsb, 5/2 lsb, and so forth). least significant bit size is given by the formula 1 lsb = fsr/4096. for the 5 v range, 1 lsb = 10 v/4096 = 2.44 mv. for the 10 v range, 1 lsb = 20 v/4096 = 4.88 mv. output coding is twos complement binary with 1 lsb = fsr/4096. the ideal input/output transfer function for the ad7864-1 is shown in tabl e 5 . figure 5. ad7864-2 analog input structure
ad7864 rev. d | page 14 of 28 table 6. ideal input/output code table for the ad7864-2 analog input 1 digital output code transition +fsr ? 3/2 lsb 2 111...110 to 111...111 +fsr ? 5/2 lsb 111...101 to 111...110 +fsr ? 7/2 lsb 111...100 to 111...101 agnd + 5/2 lsb 000...010 to 000...011 agnd + 3/2 lsb 000...001 to 000...010 agnd + 1/2 lsb 000...000 to 000...001 1 fsr is the full-scale range and is 0 v to 2.5 v and 0 v to 5 v for the ad7864-2 with v ref = 2.5 v. 2 1 lsb = fsr/4096 and is 0.61 mv (0 v to 2. 5 v) and 1.22 mv (0 v to 5 v) for the ad7864-2 with v ref = 2.5 v. ad7864-3 figure 6 shows the analog input section of the ad7864-3. the analog input range is 2.5 v on the v in1a input. the v in1b input can be left unconnected, but if it is connected to a potential, that potential must be agnd. t/h r2 ad7864-3 r1 2.5v reference to adc reference circuitry to internal comparator 6k ? v in1b v in1a v ref 01341-006 figure 6. ad7864-3 analog input structure for the ad7864-3, r1 = 6 k and r2 = 6 k. as a result, drive the v in1a input from a low impedance source. the resistor input stage is followed by the high input impedance stage of the track- and-hold amplifier. the designed code transitions take place midway between successive integer least significant bit values (that is, 1/2 lsb, 3/2 lsb, 5/2 lsb, and so on). least significant bit size is given by the formula 1 lsb = fsr/4096. output coding is twos comple- ment binary with 1 lsb = fsr/4096 = 5 v/4096 = 1.22 mv. the ideal input/ output transfer function for the ad7864-3 is shown in table 7 . table 7. ideal input/output code table for the ad7864-3 analog input 1 digital output code transition +fsr/2 ? 3/2 lsb 2 011...110 to 011...111 +fsr/2 ? 5/2 lsb 011...101 to 011...110 +fsr/2 ? 7/2 lsb 011...100 to 011...101 agnd + 3/2 lsb 000...001 to 000...010 agnd + 1/2 lsb 000...000 to 000...001 agnd ? 1/2 lsb 111...111 to 000...000 agnd ? 3/2 lsb 111...110 to 111...111 ?fsr/2 + 5/2 lsb 100...010 to 100...011 ?fsr/2 + 3/2 lsb 100...001 to 100...010 ?fsr/2 + 1/2 lsb 100...000 to 100...001 1 fsr is the full-scale range and is 5 v, with v ref = 2.5 v. 2 1 lsb = fsr/4096 = 1.22 mv (2.5 v ? ad7864-3) with v ref = 2.5 v.
ad7864 rev. d | page 15 of 28 selecting a conversion sequence any subset of the four channels, v in1 to v in4 , can be selected for conversion. the selected channels are converted in ascending order. for example, if the channel selection includes v in4 , v in1 , and v in3 , the conversion sequence is v in1 , v in3 , and then v in4 . the conversion sequence selection can be made either by using the hardware channel select input pins (sl1 through sl4) or by programming the channel select register. a logic high on a hardware channel select pin (or logic 1 in the channel select register) when convst goes logic high marks the associated analog input channel for inclusion in the conversion sequence. timing and control reading between each conversion in the conversion sequence figure 9 shows the timing and control sequence required to obtain the optimum throughput rate from the ad7864. to obtain the optimum throughput from the ad7864, the user must read the result of each conversion as it becomes available. the timing diagram in figure 9 shows a read operation each time the eoc signal goes logic low. the timing in shows a conversion on all four analog channels (sl1 to sl4 = 1, see the section), thus there are four figure 9 selecting a conversion sequence eoc pulses and four read operations to access the result of each of the four conversions. figure 7 shows the arrangement used. the h /s sel controls a multiplexer that selects the source of the conversion sequence information, that is, from the hardware channel select pins (sl1 to sl4) or from the channel selection register. when a conver- sion begins, the output from the multiplexer is latched until the end of the conversion sequence. the data bus bits, db0 to db3, (db0 representing channel 1 through db3 representing channel 4) are bidirectional and become inputs to the channel select register when rd is logic high and cs and wr are logic low. the logic state on db0 to db3 is latched into the channel select register when wr goes logic high. a conversion is initiated on the rising edge of convst . this places all four track-and-holds into hold simultaneously. new data from this conversion sequence is available for the first channel selected (v in1 ) 1.65 s later. the conversion on each subsequent channel is completed at 1.65 s intervals. the end of each conversion is indicated by the falling edge of the eoc signal. the busy output signal indicates the end-of-conversion for all selected channels (four in this case). multiplexer data bus d0 d1d2d3 cs wr channel select register sl1 sl2 sl3 sl4 hardware channel select pins h/s sel latch sequencer select individual track-and-holds for conversion transparent while waiting for convst. latched on the rising edge of convst and during a conversion sequence. 0 1341-007 wr data is read from the part via a 12-bit parallel data bus with standard cs and rd signals. the cs and rd inputs are internally gated to enable the conversion result onto the data bus. the data lines (db0 to db11) leave their high impedance state when both cs and rd are logic low. therefore, cs can be permanently tied logic low and the rd signal used to access the conversion result. because each conversion result is latched into its output data register prior to eoc going logic low, another option is to tie the eoc and rd pins together and use the rising edge of eoc to latch the conversion result. although the ad7864 has some special features that permit reading during a conversion (such as a separate supply for the output data drivers, v drive ) for optimum performance it is recommended that the read operation be completed when eoc is logic low, that is, before the start of the next conversion. although shows the read operation occurring during the figure 10 eoc pulse, a read operation can occur at any time. shows a timing specification referred to as the quiet time. quiet time is the amount of time that should be left after a read operation and before the next conversion is initiated. the quiet time depends heavily on data bus capacitance, but 50 ns to 100 ns is typical. figure 10 figure 7. channel select inputs and registers rd wr cs data t 16 t 17 t 14 t 15 t 13 data in 01341-008 the signal labeled frstdata (first data-word) indicates to the user that the pointer associated with the output data registers is pointing to the first conversion result by going logic high. the pointer is reset to point to the first data location (that is, the first conversion result,) at the end of the first conversion (frstdata figure 8. channel select ion via software control
ad7864 rev. d | page 16 of 28 logic high). the pointer is incremented to point to the next register (next conversion result) when that conversion result is available. thus, frstdata in figure 9 is shown as going low just prior to the second eoc pulse. repeated read operations during a conversion continue to access the data at the current pointer location until the pointer is incremented at the end of that conversion. note that frstdata has an indeterminate logic state after initial power-up. this means that for the first conversion sequence after power-up, the frstdata logic output may already be logic high before the end of the first conversion (this condition is indicated by the dashed line in ). also, the frstdata logic output may already be high as a result of the previous read sequence, as is the case after the fourth read in . the fourth read (rising edge of figure 9 figure 9 rd ) resets the pointer to the first data location. therefore, frstdata is already high when the next conversion sequence initiates. see the section. accessing the output data registers reading after the conversion sequence figure 10 shows the same conversion sequence as figure 9 . in this case, however, the results of the four conversions (on v in1 to v in4 ) are read after all conversions have finished, that is, when busy goes logic low. the frstdata signal goes logic high at the end of the first conversion just prior to eoc going logic low. as mentioned previously, frstdata has an indeterminate state after initial power-up, therefore frstdata may already be logic high. unlike the case when reading between each conversion, the output data register pointer is incremented on the rising edge of rd because the next conversion result is available. this means frstdata goes logic low after the first rising edge on rd . t busy quiet time t 1 t 8 t 12 t 3 t 4 t 5 t 6 t 7 v in1 v in2 v in3 v in4 100ns 100ns data convst busy eoc frstdata rd cs h/s sel sl1 to sl4 t 2 t conv t conv t conv t acq t 11 t 10 01341-009 t conv figure 9. timing diagram for reading during conversion t 10 t 8 t 4 t 3 t 6 t 1 quiet time data convst busy eoc frstdat a rd cs v in1 v in2 v in3 v in4 v in1 t busy t 2 t 10 t 7 0 1341-010 t 3 figure 10. timing diagram, reading after the conversion sequence
ad7864 rev. d | page 17 of 28 successive read operations access the remaining conversion results in an ascending channel order. each read operation increments the output data register pointer. the read operation that accesses the last conversion result causes the output data register pointer to be reset so that the next read operation accesses the first conversion result again. this is shown in figure 10 , wherein the fifth read after busy goes low accessing the result of the conversion on v in1 . thus, the output data registers act as a circular buffer in which the conversion results are continually accessible. the frstdata signal goes high when the first conversion result is available. data is enabled onto the data bus (db0 to db11) using cs and rd . both cs and rd have the same functionality as described in the previous section. there are no restrictions or performance implications associated with the position of the read operations after busy goes low. the only restriction is that there is minimum time between read operations. notice that the quiet time must be allowed before the start of the next conversion. using an external clock the logic input int /ext clk allows the user to operate the ad7864 using the internal clock oscillator or an external clock. to achieve optimum performance on the ad7864, use the internal clock. the highest external clock frequency allowed is 5 mhz. this means a conversion time of 2.6 s compared to 1.65 s when using the internal clock. in some instances, however, it may be useful to use an external clock when high throughput rates are not required. for example, two or more ad7864s can be synchronized by using the same external clock for all devices. in this way, there is no latency between output logic signals like eoc due to differences in the frequency of the internal clock oscillators. shows how the various logic outputs are synchronized to the clk signal. each conversion requires 14 clocks. the output data register pointer is reset to point to the first register location on the falling edge of the 12th clock cycle of the first conversion in the conversion sequence see the section. at this point, the logic output frstdata goes logic high. the result of the first conversion transfers to the output data registers on the falling edge of the 13th clock cycle. the frstdata signal is reset on the falling edge of the 13th clock cycle of the next conversion, that is, when the result of the second conversion is transferred to its output data register. as mentioned previously, the pointer is incremented by the rising edge of the figure 11 accessing the output data registers rd signal if the result of the next conversion is available. the eoc signal goes logic low on the falling edge of the 13th clock cycle and is reset high again on the falling edge of the 14th clock cycle. convst busy eoc rd 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 234567891011121314 1 21314 clk frstdata first conversion complete last conversion complete 01341-011 figure 11. using an external clock
ad7864 rev. d | page 18 of 28 standby mode operation the ad7864 has a standby mode whereby the device can be placed in a low current consumption mode (5 a typical). the ad7864 is placed in standby by bringing the logic input stby low. the ad7864 can be powered up again for normal opera- tion by bringing stby logic high. the output data buffers remain operational while the ad7864 is in standby. this means the user can continue to access the conversion results while the ad7864 is in standby. this feature can be used to reduce the average power consumption in a system using low throughput rates. to reduce average power consumption, the ad7864 can be placed in standby at the end of each conversion sequence, that is, when busy goes low and is taken out of standby again prior to the start of the next conversion sequence. the time it takes the ad7864 to come out of standby is referred to as the wake-up time. the wake-up time limits the maximum throughput rate at which the ad7864 can be operated when powering down between conver- sion sequences. the ad7864 wakes up in approximately 2 s when using an external reference. the wake-up time is also 2 s when the standby time is less than 1 ms while using the internal refer- ence. shows the wake-up time of the ad7864 for standby times greater than 1 ms. note that when the ad7864 is left in standby for periods of time greater than 1 ms, the part requires more than 2 s to wake up. for example, after initial power-up using the internal reference, the ad7864 requires 6 ms to power up. the maximum throughput rate that can be achieved when powering down between conversions is 1/(t busy + 2 s) = 100 ksps, approximately. when operating the ad7864 in a standby mode between conversions, the power savings can be significant. for example, with a throughput rate of 10 ksps, the ad7864 is powered down (i dd = 5 a) for 90 s out of every 100 s (see ). figure 12 figure 13 therefore, the average power consumption drops to 125/10 mw or 12.5 mw approximately. standby time (seconds) 1.0 0.9 0 10 0.0001 0.001 0.01 0.1 1 0.6 0.3 0.2 0.1 0.8 0.7 0.4 0.5 power-up time (ms) +105c +25c ?40c 01341-012 figure 12. power-up time vs. standby time using the on-chip reference (decoupled with 0.1 f capacitor) accessing the output data registers there are four output data registers, one for each of the four possible conversion results from a conversion sequence. the result of the first conversion in a conversion sequence is placed in register 1, the second result is placed in register 2, and so forth. for example, if the conversion sequence v in1 , v in3 , and v in4 is selected (see the selecting a conversion sequence section), the results of the conversion on v in1 , v in3 , and v in4 are placed in register 1 to register 3, respectively. the output data register pointer is reset to point to register 1 at the end of the first con- version in the sequence, immediately prior to eoc going low. at this point, the logic output, frstdata, goes logic high to indicate that the output data register pointer is addressing reg- ister 1. when cs and rd are both logic low, the contents of the addressed register are enabled onto the data bus (db0 to db11). convst busy stby 100s 7s t busy i dd = 20a t busy 2s t wake-up 01341-013 figure 13. power-down between conversion sequences
ad7864 rev. d | page 19 of 28 when reading the output data registers after a conversion sequence, that is, when busy goes low, the register pointer is incremented on the rising edge of the rd signal, as shown in . however, when reading the conversion results during the conversion sequence, the pointer is not incremented until a valid conversion result is in the register to be addressed. in this case, the pointer is incremented when the conversion has ended and the result has been transferred to the output data register. this happens immediately before figure 14 eoc goes low, therefore eoc may be used to enable the register contents onto the data bus, as described in the subsection within the section. the pointer is reset to point to register 1 on the rising edge of the reading between each conversion in the conversion sequence selecting a conversion sequence rd signal when the last conversion result in the sequence is being read. in the example shown, this means that the pointer is set to register 1 when the contents of register 3 are read. db0 to db11 output drivers oe no. 1 not valid (v in3 ) (v in1 ) (v in4 ) oe no. 2 oe no. 3 oe no. 4 2-bit counter v drive oe rd cs reset decode output data registers *the pointer is not incremented by a rising edge on rd until the conversion result is in the output data register. the pointer is reset when the last conversion result is read. frstdat a pointer* ad7864 0 1341-014 figure 14. output data registers
ad7864 rev. d | page 20 of 28 offset and full-scale adjustment in most digital signal processing (dsp) applications, offset and full-scale errors have little or no effect on system performance. offset error can always be eliminated in the analog domain by ac coupling. full-scale error effect is linear and does not cause problems as long as the input signal is within the full dynamic range of the adc. invariably, some applications require that the input signal spans the full analog input dynamic range. in such applications, offset and full-scale error have to be adjusted to zero. figure 15 shows a circuit that can be used to adjust the offset and full-scale errors on the ad7864 (v inxa on the ad7864-1 version is shown for example purposes only). where adjustment is required, offset error must be adjusted before full-scale error. this is achieved by trimming the offset of the op amp driving the analog input of the ad7864 while the input voltage is 1/2 lsb below analog ground. the trim procedure is as follows: apply a voltage of ?2.44 mv (?1/2 lsb) at v 1 in figure 15 and adjust the op amp offset voltage until the adc output code flickers between 1111 1111 1111 and 0000 0000 0000. adjust gain error at either the first code transition (adc negative full scale) or the last code transition (adc positive full scale). the trim procedures for both cases are as follows. positive full-scale adjust apply a voltage of 9.9927 v (fs ? 3/2 lsb) at v 1 and adjust r2 until the adc output code flickers between 0111 1111 1110 and 0111 1111 1111. negative full-scale adjust apply a voltage of ?9.9976 v (?fs + 1/2 lsb) at v 1 and adjust r2 until the adc output code flickers between 1000 0000 0000 and 1000 0000 0001. an alternative scheme for adjusting full-scale error in systems that use an external reference is to adjust the voltage at the v ref pin until the full-scale error for any of the channels is adjusted out. good full-scale matching of the channels ensures small full-scale errors on the other channels. v 1 r1 10k? r2 500? r3 10k? agnd ad7864-1* *additional pins omitted for clarity. input range = 10v 10k ? r5 10k? r4 v inxa 01341-015 figure 15. full-scale adjust circuit
ad7864 rev. d | page 21 of 28 dynamic specifications the ad7864 is specified and 100% tested for dynamic perfor- mance specifications as well as traditional dc specifications, such as integral and differential nonlinearity. these ac specifications are required for signal processing applications such as phased array sonar, adaptive filters, and spectrum analysis. these applications require information on the effect of the adc on the spectral content of the input signal. thus, the parameters for which the ad7864 is specified include snr, harmonic distortion, inter- modulation distortion, and peak harmonics. these terms are discussed in more detail in the following sections. signal-to-noise ratio (snr) snr is the measured signal-to-noise ratio at the output of the adc. the signal is the rms magnitude of the fundamental. noise is the rms sum of all the nonfundamental signals up to half of the sampling frequency (f s /2) excluding dc. snr depends on the number of quantization levels used in the digitization process; the more levels, the smaller the quantization noise. the theoretical signal-to-noise ratio for a sine wave input is given by snr = (6.02 n + 1.76) db (1) where n is the number of bits. thus, for an ideal 12-bit converter, snr = 74 db. figure 16 shows a histogram plot for 8192 conversions of a dc input using the ad7864 with a 5 v supply. the analog input was set at the center of a code. the figure shows that all the codes appear in the one output bin, indicating very good noise performance from the adc. adc code 1054 1 0 5 5 1 0 5 6 1 0 5 7 1 0 5 8 1 0 5 9 1 0 6 0 1 0 6 1 1 0 6 2 1 0 6 3 1 0 6 4 9000 8000 0 4000 3000 2000 1000 6000 5000 7000 counts 01341-016 figure 16. histogram of 8192 conversions of a dc input the output spectrum from the adc is evaluated by applying a sine wave signal of very low distortion to the analog input. a fast fourier transform (fft) plot is generated from which the snr data can be obtained. figure 17 shows a typical 4096 point fft plot of the ad7864 with an input signal of 99.9 khz and a sampling frequency of 500 khz. the snr obtained from this graph is 72.6 db. note that the harmonics are taken into account when calculating the snr. 0 50 100 ?110 ?100 ?90 ?80 ?70 ?60 ?50 ?40 ?30 ?20 ?10 0 150 frequency (khz) (db) 200 250 ad7864-1 @ 25c 5v supply sampling at 499,712hz input frequency of 99,857hz 8192 samples taken 01341-017 figure 17. fft plot effective number of bits the formula given in equation 1 relates the snr to the number of bits. rewriting the formula, as in equation 2, it is possible to get a measure of performance expressed in effective number of bits (n). 02.6 76.1 ? = snr n (2) the effective number of bits for a device can be calculated directly from its measured snr. figure 18 shows a typical plot of effective number of bits vs. frequency for an ad7864-2. frequency (khz) 12 4 effective numbers of bits 11 8 7 6 5 10 9 0 3000 500 ?40c +25c +105c 1000 1500 2000 2500 01341-018 figure 18. effective numbers of bits vs. frequency intermodulation distortion with inputs consisting of sine waves at two frequencies, fa and fb, any active device with nonlinearities creates distortion products at sum and difference frequencies of mfa nfb where m, n = 0, 1, 2, 3, and so forth. intermodulation terms are those for which neither m nor n are equal to zero. for example, the second-order
ad7864 rev. d | page 22 of 28 terms include (fa + fb) and (fa ? fb), whereas the third-order terms include (2fa + fb), (2fa ? fb), (fa + 2fb), and (fa ? 2fb). using the ccif standard where two input frequencies near the top end of the input bandwidth are used, the second- and third- order terms are of different significance. the second-order terms are usually distanced in frequency from the original sine waves, whereas the third-order terms are usually at a frequency close to the input frequencies. as a result, the second- and third-order terms are specified separately. the calculation of the intermodulation distortion is as per the thd specification where it is the ratio of the rms sum of the individual distortion products to the rms amplitude of the fundamental expressed in decibels. in this case, the input consists of two, equal amplitude, low distortion sine waves. figure 19 shows a typical imd plot for the ad7864. 01020 ?100 ?90 ?80 ?70 ?60 ?50 ?40 ?30 ?20 ?10 0 30 frequency (khz) (db) 40 60 50 ad7864-1 @ 25c 5v supply sampling at 131072hz input frequency of 48,928hz and 50,016hz 4096 samples taken 0 1341-019 figure 19. imd plot ac linearity plots the plots shown in figure 20 and figure 21 show typical dnl and inl plots for the ad7864. 0 500 1000 1500 2000 2500 3000 3500 adc code 3 2 1 0 ?1 ?2 ?3 dnl (lsb) 4000 01341-020 figure 20. typical dnl plot ?0.5 0 0.5 1.0 1.5 2.0 2.5 ?2.5 ?2.0 ?1.5 ?1.0 0 500 1000 1500 2000 2500 3000 3500 4000 adc code inl (lsb) 0 1341-021 figure 21. typical inl plot measuring aperture jitter a convenient way to measure aperture jitter is to use the relationship it is known to have with snr (signal-to-noise plus distortion) given as follows: () ? ? ? ? ? ? ? ? = f snr in jitter 2 1 log20 10 (3) where: snr jitter is the signal-to-noise due to the rms time jitter. is the rms time jitter. f in is the sinusoidal input frequency (1 mhz in this case). equation 3 demonstrates that the signal-to-noise ratio due to jitter degrades significantly with frequency. at low input fre- quencies, the measured snr performance of the ad7864 is indicative of noise performance due to quantization noise and system noise only (72 db used as a typical figure in this example). therefore, by measuring the overall snr performance (including noise due to jitter, system, and quantization) of the ad7864, a good estimation of the jitter performance of the ad7864 can be calculated. frequency (hz) 12 11 5 9 8 7 6 10 enob 900k 950k 1.00m 1.05m 1.10m 01341-022 figure 22. enob of the ad7864 at 1 mhz
ad7864 rev. d | page 23 of 28 from figure 22 , the enob of the ad7864 at 1 mhz is approximately 11 bits. this is equivalent to 68 db snr. snr total = snr jitter + snr quant = 68 db 68 db = snr jitter + 72 db (at 100 khz) snr jitter = 70.2 db from equation 3 70.2 db = 20 log 10 [1/(2 1 mhz )] = 49 ps where is the rms jitter of the ad7864.
ad7864 rev. d | page 24 of 28 microprocessor interfacing the high speed parallel interface of the ad7864 allows easy interfacing to most dsps and microprocessors. this interface consists of the data lines (db0 to db11), cs , rd , wr , eoc , and busy. ad7864 to adsp-2100/ adsp-2101/adsp-2102 interface figure 23 shows an interface between the ad7864 and the adsp-210x. the convst signal can be generated by the adsp-210x or from some other external source. shows the figure 23 cs being generated by a combination of the dms signal and the address bus of the adsp-210x. in this way, the ad7864 is mapped into the data memory space of the adsp-210x. the ad7864 busy line provides an interrupt to the adsp-210x when the conversion sequence is complete on all the selected channels. the conversion results can then be read from the ad7864 using successive read operations. alternately, one can use the eoc pulse to interrupt the adsp-210x when the conversion on each channel is complete when reading between each conversion in the conversion sequence ( ). the ad7864 is read using the following instruction: figure 9 mr0 = dm(adc) where mr0 is the adsp-210x mr0 register and adc is the ad7864 address. cs rd wr busy convst db0 to db11 ad7864 v in1 v in2 v in3 v in4 dt1/f0 irqn rd wr d0 to d24 dms a0 to a13 adsp-210x address decode 01341-023 figure 23. ad7864 to adsp-210x interface ad7864 to tms320c5x interface figure 24 shows an interface between the ad7864 and the tms320c5x. as with the previous interfaces, conversion can be initiated from the tms320c5x or from an external source, and the processor is interrupted when the conversion sequence is completed. the cs signal to the ad7864 is derived from the ds signal and a decode of the address bus. this maps the ad7864 into external data memory. the rd signal from the tms320c5x is used to enable the adc data onto the data bus. the ad7864 has a fast parallel bus, consequently there are no wait state requirements. the following instruction is used to read the conversion results from the ad7864: in d,adc where d is the data memory address and adc is the ad7864 address. cs rd wr busy convst db0 to db11 ad7864 v in1 v in2 v in3 v in4 pa0 intn rd we d0 to d15 ds a0 to a13 tms320c5x address decode 01341-024 figure 24. ad7864 to tms320c5x interface ad7864 to mc68hc000 interface an interface between the ad7864 and the mc68hc000 is shown in figure 25 . the conversion can be initiated from the mc68hc000 or from an external source. the ad 7864 busy line can be used to interrupt the processor or, alternatively, software delays can ensure that the conversion has been completed before a read to the ad7864 is attempted. because of the nature of its interrupts, the mc68hc000 requires additional logic (not shown in figure 25 ) to allow it to be interrupted correctly. for further information on mc68hc000 interrupts, consult the addendum to mc68000 users manual . the mc68hc000 as and r/ w outputs are used to generate a separate rd input signal for the ad7864. rd is used to drive the mc68hc000 dtack input to allow the processor to execute a normal read operation to the ad7864. the conversion results are read using the following mc68hc000 instruction: move.w adc,d0 where d0 is the mc68hc000 d0 register and adc is the ad7864 address.
ad7864 rev. d | page 25 of 28 cs rd convst db0 to db11 ad7864 v in1 v in2 v in3 v in4 dtack as d0 to d15 a0 to a15 mc68hc000 address decode clock r/w 01341-025 figure 25. ad7864 to mc68hc000 interface vector motor control the current drawn by a motor can be split into two components: one produces torque and the other produces magnetic flux. for optimal performance of the motor, control these two compo- nents independently. in conventional methods of controlling a three-phase motor, the current (or voltage) supplied to the motor and the frequency of the drive are the basic control variables. however, both the torque and flux are functions of current (or voltage) and frequency. this coupling effect can reduce the performance of the motor because, for example, if the torque is increased by increasing the frequency, the flux tends to decrease. vector control of an ac motor involves controlling phase in addition to drive and current frequency. controlling the phase of the motor requires feedback information on the position of the rotor relative to the rotating magnetic field in the motor. using this information, a vector controller mathematically trans- forms the three-phase drive currents into separate torque and flux components. the ad7864, with its 4-channel simultaneous sampling capability, is ideally suited for use in vector motor control applications. a block diagram of a vector motor control application using the ad7864 is shown in figure 26 . the position of the field is derived by determining the current in each phase of the motor. only two phase currents need to be measured because the third can be calculated if two phases are known. v in1 and v in2 of the ad7864 are used to digitize this information. simultaneous sampling is critical to maintain the relative phase information between the two channels. a current sensing isola- tion amplifier, transformer, or hall effect sensor is used between the motor and the ad7864. rotor information is obtained by measuring the voltage from two of the inputs to the motor. v in3 and v in4 of the ad7864 are used to obtain this information. once again, the relative phase of the two channels is important. a dsp microprocessor is used to perform the mathematical transformations and control loop calculations on the information fed back by the ad7864. dac torque and flux control loop calculations and two to three phase information dsp microprocessor dac dac drive circuitry i c i b i a v b v a transformation to torque and flux current components + ? + ? ad7864* v in1 v in2 v in3 v in4 isolation amplifiers voltage attenuators torque setpoint flux setpoint *additional pins omitted for clarity. three- phase motor 01341-027 figure 26. vector motor control using the ad7864
ad7864 rev. d | page 26 of 28 multiple ad7864s in a system figure 27 shows a system where a number of ad7864s are configured to handle multiple input channels. this type of configuration is common in applications such as sonar and radar. the ad7864 is specified with maximum limits on aperture delay match. this means that the user knows the difference in the sampling instant between all channels. this allows the user to maintain relative phase information between the different channels. the ad7864 has a maximum aperture delay matching of 4 ns. all ad7864s use the same external sar clock (5 mhz). therefore, the conversion time for all devices is identical; consequently, all devices can be read simultaneously. in the example shown in figure 27 , the data outputs of two ad7864s are enabled onto a 32-bit wide data bus when eoc goes low. eoc ad7864 v in1 v in2 v in3 v in4 address decode v ref clkin cs rd 12 32 ad7864 v in1 v in2 v in3 v in4 v ref clkin 12 adsp-2106x rd ref193 5mhz cs rd 01341-026 figure 27. multiple ad7864s in multichannel system
ad7864 rev. d | page 27 of 28 outline dimensions compliant to jedec standards mo-112-aa-1 041807-a 14.15 13.90 sq 13.65 0.45 0.30 2.45 max 1.03 0.88 0.73 top view (pins down) 12 44 1 22 23 34 33 11 0.25 min 2.20 2.00 1.80 7 0 view a rotated 90 ccw 0.23 0.11 10.20 10.00 sq 9.80 0.80 bsc lead pitch lead width 0.10 coplanarity v i e w a s e a t i n g p l a n e 1 . 9 5 r e f pin 1 figure 28. 44-lead metric quad flat package [mqfp] (s-44-2) dimensions shown in millimeters ordering guide model input ranges relative accuracy temperature range 1 package description package option ad7864asz-1 2 5 v, 10 v 1 lsb ?40c to +85c 44-lead mqfp s-44-2 ad7864asz-1reel 2 5 v, 10 v 1 lsb ?40c to +85c 44-lead mqfp s-44-2 ad7864bsz-1 2 5 v, 10 v 0.5 lsb ?40c to +85c 44-lead mqfp s-44-2 ad7864bsz-1reel 2 5 v, 10 v 0.5 lsb ?40c to +85c 44-lead mqfp s-44-2 ad7864asz-2 2 0 v to 2.5 v, 0 v to 5 v 1 lsb ?40c to +85c 44-lead mqfp s-44-2 ad7864asz-2reel 2 0 v to 2.5 v, 0 v to 5 v 1 lsb ?40c to +85c 44-lead mqfp s-44-2 ad7864asz-3 2 2.5 v 1 lsb ?40c to +85c 44-lead mqfp s-44-2 ad7864asz-3reel 2 2.5 v 1 lsb ?40c to +85c 44-lead mqfp s-44-2 eval-ad7864-2cb 3 evaluation board eval-ad7864-3cb 3 evaluation board eval-control brd2 4 controller board 1 the a version is fully specified up to 1 05c with a maximum sample rate of 450 ksps and i dd maximum (normal mode) of 26 ma. 2 z = rohs compliant part. 3 this can be used as a stand alone evaluation board or in co njunction with the evaluation controller board for evaluation/demon stration purposes. 4 this board is a complete unit, allowing a pc to control and communicate with all analog devices, inc., evaluation boards endin g in the cb designators. to order a complete evaluation kit, the pa rticular adc evaluation board ne eds to be ordered, for example, eval-ad7864-1cb, the eval-contro l brd2, and a 12 v ac transformer. see the evaluation board a pplication note for more information.
ad7864 rev. d | page 28 of 28 notes ?1998C2009 analog devices, inc. all rights reserved. trademarks and registered trademarks are the prop erty of their respective owners. d01341-0-2/09(d)

▲Up To Search▲

Price & Availability of EVAL-AD7864-3CB3

	To Download EVAL-AD7864-3CB3 Datasheet File
If you can't view the Datasheet, Please click here to try to view without PDF Reader .