View WM2638_3464962.PDF datasheet online --- IC-ON-LINE

Datasheet File OCR Text:

WM2638 dual 12-bit serial input voltage output dac with internal reference production data, rev 1.0, july 1999 wolfson microelectronics ltd lutton court, bernard terrace, edinburgh, eh8 9nx, uk tel: +44 (0) 131 667 9386 fax: +44 (0) 131 667 5176 email: sales@wolfson.co.uk http://www.wolfson.co.uk production data contains final specifications current on publication date. supply of products conforms to wolfson microelectronics' terms and conditions. last printed 15/07/99 15:58 1999 wolfson microelectronics ltd . features two 12-bit voltage output dacs single 2.7v to 5.5v supply operation dnl 0.4 lsbs, inl 2 lsbs low power consumption 12mw typical in fast mode 6mw typical in slow mode tms320, (q)spi ? ? , and microwire ? ? compatible serial interface programmable settling time 1 m m s or 3.5 m m s typical power down mode 10na applications battery powered test instruments digital offset and gain adjustment battery operated/remote industrial controls machine and motion control devices cellular telephones wireless telephones and communication systems speech synthesis arbitrary waveform generation ordering information device temp. range package WM2638cd 0 to 70c 8-pin soic WM2638id -40 to 85c 8-pin soic description the WM2638 is a dual 12-bit voltage output, resistor string, digital- to-analogue converter. a software controlled power down mode is provided that reduces current consumption to 10na. the WM2638 features an internal programmable voltage reference simplifying overall system design. a reference voltage may also be supplied externally. the device has been designed to interface efficiently to industry standard microprocessors and dsps, including the tms320 family. the WM2638 is programmed with a 16-bit serial word comprising of a latch address, mode control bits and dac or control data. excellent performance is delivered with a typical dnl of 0.4lsbs. the programmable settling time allows the designer to optimise speed versus power consumption. the output stage is buffered by a x2 gain near rail-to-rail amplifier. the device is available in an 8-pin soic package ideal for space- critical applications. commercial temperature (0 to 70c) and industrial temperature (-40 to 85c) variants are supported. block diagram typica l performance (7) outb (4) outa 12-bit dac b holding latch 12-bit dac a latch ref(6) power-on reset din (1) sclk (2) ncs (3) (5) agnd powerdown/ speed control 12-bit dac b latch dac output buffer 2-bit control latch x1 x2 x2 dac output buffer reference input buffer reference input buffer 16-bit shift register and control logic 2-bit reference select latch 1.024v/2.048v selectable reference x1 v dd (8) c x1 reference output buffer with ouput enable WM2638 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 0 512 1024 1536 2048 2559 3071 3583 4095 digital code dnl- lsb 5v = vdd, v ref = external, speed = fast mode, load = 10k/100pf
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 2 pin configuration 1 2 3 4 ncs din sclk agnd ref outb vdd outa 5 6 7 8 pin description pin no name type description 1 din digital input serial data input. 2 sclk digital input serial clock input. 3 ncs digital input chip select. this pin is active low. 4 outa analogue output dac a analogue output. 5 agnd supply analogue ground. 6 ref analogue i/o reference voltage input/output. 7 outb analogue output dac b analogue output 8 vdd supply positive power supply. absolute maximum ratings absolute maximum ratings are stress ratings only. permanent damage to the device may be caused by continuously operating at or beyond these limits. device functional operating limits and guaranteed performance specifications are given under electrical characteristics at the test conditions specified esd sensitive device. this device is manufactured on a cmos process. it is therefore generically susceptible to damage from excessive static voltages. proper esd precautions must be taken during handling and storage of this device. condition min max supply voltage, vdd to agnd 7v digital input voltage range to agnd -0.3v vdd + 0.3v reference input voltage range to agnd -0.3v vdd + 0.3v output voltage at out from external source vdd + 0.3v operating temperature range, t a WM2638c WM2638i 0 c -40 c 70 c 85 c storage temperature -65 c 150 c lead temperature 1.6mm (1/16 inch) soldering for 10 seconds 260 c recommended operating conditions parameter symbol test conditions min typ max unit supply voltage vdd 2.7 5.5 v high-level digital input voltage v ih 2 0.8 v low-level digital input voltage v il v reference voltage to ref v ref vdd - 1.5 v load resistance r l 2 k w load capacitance c l 100 serial clock rate f sclk 20 WM2638c 0 70 c operating free-air temperature t a WM2638i -40 85 c note : reference voltages greater than vdd/2 will cause saturation for large dac codes.
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 3 electrical characteristics test conditions: r l = 10k w , c l = 100pf. vdd = 5v 10%, v ref = 2.048v and vdd = 3v 10%, v ref = 1.024v over recommended operating free-air temperature range (unless noted otherwise) parameter symbo l test conditions min typ max unit static dac specifications resolution 12 bits integral non-linearity inl see note 1 2.0 4.0 lsb differential non-linearity dnl see note 2 0.4 1 lsb zero code error zce see note 3 24 mv gain error ge see note 4 0.6 % fsr d.c. power supply rejection ratio dc psrr see note 5 0.5 mv/v zero code error temperature coefficient see note 6 10 ppm/ c gain error temperature coefficient see note 6 10 ppm/ c dac output specifications output voltage range 0 vdd - 0.4 v output load regulation 2k w to 10k w load see note 7 0.1 0.25 % power supplies active supply current i dd no load, v ih = vdd, v il = 0v vdd = 5v, v ref = 2.048v, internal slow fast vdd = 5v, v ref = 2.048v, external slow fast vdd = 3v, v ref = 1.024v, internal slow fast vdd = 3v, v ref = 1.024v, external slow fast see note 8 2.2 4.3 1.8 3.9 1.8 3.8 1.5 3.5 2.7 5.2 2.2 4.8 2.3 4.7 1.9 4.3 ma ma ma ma ma ma ma ma power down supply current no load, all digital inputs 0v or vdd, see note 9 0.01 10 m a dynamic dac specifications slew rate dac code 32 to 4095, 10%-90% slow fast see note 10 1.5 8.0 v/ m s v/ m s settling time dac code 32 to 4095 slow fast see note 11 3.5 1.0 m s m s glitch energy code 2047 to 2048 10 nv-s signal to noise ratio snr fs = 400ksps, f out = 1khz, bw = 20khz see note 12 69 74 db signal to noise and distortion ratio snrd fs = 400ksps, f out = 1khz, bw = 20khz see note 12 58 67 db
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 4 test conditions: r l = 10k w , c l = 100pf. vdd = 5v 10%, v ref = 2.048v and vdd = 3v 10%, v ref = 1.024v over recommended operating free-air temperature range (unless noted otherwise) parameter symbo l test conditions min typ max unit total harmonic distortion thd fs = 400ksps, f out = 1khz, bw = 20khz see note 12 -69 -57 db spurious free dynamic range spfdr fs = 400ksps, f out = 1khz, bw = 20khz see note 12 57 72 db reference configured as input reference input resistance r refin 10 m w reference input capacitance c refin 55 pf reference feedthrough v ref = 1v pp at 1khz + 1.024v dc, dac code 0 -60 db reference input bandwidth v ref = 0.2v pp + 1.024v dc dac code 2048 slow fast 1.0 1.0 mhz mhz reference configured as output low reference voltage v refoutl 1.003 1.024 1.045 v high reference voltage v refouth vdd > 4.75v 2.027 2.048 2.069 v output source current i refsrc 1 ma output sink current i refsnk -1 ma load capacitance 100 pf psrr -48 db digital inputs high level input current i ih input voltage = vdd 1 m a low level input current i il input voltage = 0v -1 m a input capacitance c i 8 pf notes: 1. integral non-linearity (inl) is the maximum deviation of the output from the line between zero and full scale (excluding the effects of zero code and full scale errors). 2. differential non-linearity (dnl) is the difference between the measured and ideal 1lsb amplitude change of any adjacent two codes. a guarantee of monotonicity means the output voltage changes in the same direction (or remains constant) as a change in digital input code. 3. zero code error is the voltage output when the dac input code is zero. 4. gain error is the deviation from the ideal full scale output excluding the effects of zero code error. 5. power supply rejection ratio is measured by varying vdd from 4.5v to 5.5v and measuring the proportion of this signal imposed on the zero code error and the gain error. 6. zero code error and gain error temperature coefficients are normalised to full scale voltage. 7. output load regulation is the difference between the output voltage at full scale with a 10k w load and 2k w load. it is expressed as a percentage of the full scale output voltage with a 10k w load. 8. i dd is measured while continuously writing code 2048 to the dac. for v ih < vdd - 0.7v and v il > 0.7v supply current will increase. 9. typical supply current in power down mode is 10na. production test limits are wider for speed of test. 10. slew rate results are for the lower value of the rising and falling edge slew rates 11. settling time is the time taken for the signal to settle to within 0.5lsb of the final measured value for both rising and falling edges. limits are ensured by design and characterisation, but are not production tested. 12. snr, snrd, thd and spfdr are measured on a synthesised sinewave at frequency f ou t generated with a sampling frequency fs.
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 5 serial interface ncs sclk din d15 d14 d13 d12 d11 d0 t sucss t wcl t wch t sucs1 t sucs2 t hdclk t sudclk figure 1 timing diagram test conditions: r l = 10k w , c l = 100pf. vdd = 5v 10%, v ref = 2.048v and vdd = 3v 10%, v ref = 1.024v over recommended operating free- air temperature range (unless noted otherwise) symbol test conditions min typ max unit t sucss setup time ncs low before sclk low 10 ns t suscs1 setup time, rising edge of sclk to rising edge of ncs, external end of write 10 ns t suscs2 setup time, rising edge of sclk to falling edge of ncs, start of next write cycle 5 ns t wcl pulse duration, sclk high 25 ns t wch pulse duration, sclk low 25 ns t sudclk setup time, data ready before sclk falling edge 10 ns t hdclk hold time, data held valid after sclk falling edge 5 ns
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 6 typical performance graphs vdd = 5v, v ref = external 2.048v, speed = fast mode, load = 10k/100pf -3 -2 -1 0 1 2 3 0 512 1024 1536 2048 2559 3071 3583 4095 digital code inl - lsb figure 2 integral non-linearity 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 1 2 3 4 5 6 7 8 9 10 isink- ma output voltage - v slow fast vdd = 3v, v ref = 1v, input code = 0 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0 1 2 3 4 5 6 7 8 9 10 isink - ma output voltage - v slow fast vdd = 5v, v ref = 2v, input code = 0 figure 3 sink current vdd = 3v figure 4 sink current vdd = 5v 2.02 2.025 2.03 2.035 2.04 2.045 2.05 2.055 2.06 0 1 2 3 4 5 6 7 8 9 10 11 isource- ma output voltage - v slow fast vdd = 3v, v ref = 1v, input code = 4095 4.06 4.065 4.07 4.075 4.08 4.085 4.09 4.095 4.1 0 1 2 3 4 5 6 7 8 9 10 11 isource - ma output voltage - v slow fast vdd = 5v, v ref = 2v, input code = 4095 figure 5 source current vdd = 3v figure 6 source current vdd = 5v
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 7 device description general function the device uses a resistor string network buffered with an op amp to convert 12-bit digital data to analogue voltage levels (see block diagram). the output voltage is determined by the reference input voltage and the input code according to the following relationship: output voltage = ( ) 4096 code v 2 ref input output 1111 1111 1111 ( ) 4096 4095 v 2 ref : : 1000 0000 0001 ( ) 4096 2049 v 2 ref 1000 0000 0000 ( ) ref ref v 4096 2048 v 2 = 0111 1111 1111 ( ) 4096 2047 v 2 ref : : 0000 0000 0001 ( ) 4096 1 v 2 ref 0000 0000 0000 0v table 1 binary code table (0v to 2v ref output), gain = 2 power on reset an internal power-on-reset circuit resets the dac registers to all 0s on power-up. buffer amplifier the output buffer has a near rail-to-rail output with short circuit protection and can reliably drive a 2k w load with a 100pf load capacitance. serial interface when chip select (ncs) is low, the input data is read into a 16-bit shift register with the input data clocked in most significant bit first. the falling edge of the sclk input shifts the data into the input register. after 16 bits have been transferred, the next rising edge on sclk or ncs then transfers the data to the dac latch. when ncs is high, input data cannot be clocked into the input register (see table 2). serial clock and update rate figure 1 shows the device timing. the maximum serial rate is: f sclk max = mhz 20 t t 1 min wcl min wch = + the digital update rate is limited to an 800ns period, or 1.25mhz frequency. however, the dac settling time to 12 bits limits the update rate for large input step transitions.
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 8 software configuration options the 16 bits of data can be transferred with the sequence shown in table 2. d11-d0 contains the 12-bit data word. d15-d12 hold the programmable options which are summarized in table 3. d15 d14 d13 d12 d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 r1 spd pwd r0 new dac value or control register value table 2 serial word format programmable settling time settling time is a software selectable 3.5 m s or 1 m s typical, to within 0.5lsb of final value. this is controlled by the value of d14. a one defines a settling time of 1 m s, a zero defines a settling time of 3.5 m s. programmable power down the power down function is controlled by d13. a zero configures the device as active, or fully powered up, a one configures the device into power down mode. when the power down function is released the device reverts back to the dac code set prior to power down. register addressing a separate internal control register is available. this is accessed from the register access bits r1 (bit d15) and r0 (bit d12). r1 (bit d15) r0 (bit d12) register 0 0 write data to dac b and buffer 0 1 write data to buffer 1 0 write data to dac a and update dac b with buffer content 1 1 write data to control register table 3 programmable options the contents of the control register, shown below in table 3, are used to program the internal reference function d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 x x x x x x x x x x ref1 ref0 table 4 control register contents programmable internal reference the reference can be sourced internally or externally under software control. if an external reference voltage is applied to the ref pin, the device must be configured to accept this. if an external reference is selected, the reference voltage input is buffered which makes the dac input resistance independent of code. the ref pin has an input resistance of 10m w and an input capacitance of typically 55pf. the reference voltage determines the dac full-scale output. if an internal reference is selected, a voltage of 1.024v or 2.048v is available. the internal reference can source up to 1ma and can therefore be used as an external system reference. ref1 (bit d1) ref0 (bit d0) register 0 0 external 0 1 1.024v 1 0 2.048v 1 1 external table 5 internal reference options
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 9 examples: 1. set internal reference voltage to 2.048v d15 d14 d13 d12 d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 1 x 0 1 x x x x x x x x x x 1 0 2. write new dac a value and update dac a output d15 d14 d13 d12 d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0 0 x 0 0 new dac a output value function of the latch control bits (d15 and d12) purpose and use of the double buffer normally only one dac output can change after a write. the double buffer allows both dac outputs to change after a single write. this is achieved by the two following steps. a double buffer only write is executed to store the new dac b data without changing the dac a and b outputs. following the previous step, a write to latch a is executed. this writes the serial interface register (sir) data to latch a and also writes the double buffer contents to latch b. thus both dacs receive their new data at the same time and so both dac outputs begin to change at the same time. unless a double buffer only write is issued, the latch b and double buffer contents are identical. thus, following a write to latch a or b with another write to latch a does not change the latch b contents. three data transfer options are possible. all transfers occur immediately after ncs goes high (or on the sixteenth positive sclk edge, whichever is earlier) and are described in the following sections). latch a write, latch b update (d15 = high, d12 = low) the serial interface register (sir) data are written to latch a and the double buffer latch contents are written to latch b. the double buffer contents are unaffected. this program bit condition allows simultaneous output updates of both dacs. serial interface register d12 = low d15 = high double buffer latch latch b to dac b latch a to dac a figure 2 latch a write, latch b update latch b and double buffer 1 write (d15 = low, d12 = low) the sir data are written to both latch b and the double buffer. latch a is unaffected. serial interface register d12 = low d15 = low double buffer latch latch b to dac b latch a to dac a figure 3 latch b and double buffer write double buffer only write (d15 = low, d12 = high) the sir data are written to the double buffer only. latch a and b contents are unaffected.
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 10 serial interface register d12 = high d15 = low double buffer latch b to dac b latch a to dac a figure 4 double buffer only write operational examples 1. changing the latch a data from zero to full code assuming that latch a starts at zero code (e.g. after power up), the latch can be filled with 1s by writing (bit d15 on the left, d0 on the right) 1x00 1111 1111 1111 to the serial interface. bit d14 can be zero to select slow mode or one to select fast mode. the latch b contents and dac b output are not changed by this write unless the double buffer contents are different from the latch b contents. this can only be true if the last write was a double buffer-only write. 2. changing the latch b data from zero to full code assuming that latch b starts at zero code (e.g. after power up), the latch can be filled with 1s by writing (bit d15 on the left, d0 on the right) 0x00 1111 1111 1111 to the serial interface. bit d14 can be zero to select slow mode or one to select fast mode. the data (bits d0 to d11) are written to both the double buffer and latch b. the latch a contents and the dac a output are not changed by this write. 3. double buffered change of both dac outputs assuming that dacs a and b start at zero code (e.g. after power up), if dac a is to be driven to mid-scale and dac b to full-scale, and if the outputs are to begin rising at the same time, this can be achieved as follows: first, 0d01 1111 1111 1111 is written (bit d15 on the left, d0 on the right) to the serial interface. this loads the full-scale code into the double buffer but does not change the latch b contents and the dac b output voltage. the latch a contents and the dac a output are also unaffected by this write operation. changing from fast to slow to fast mode changes the supply current which can glitch the outputs, and so d14 (designated by d in the above data word) should be set to maintain the speed mode set by the previous write. next, 1d00 1000 0000 0000 is written (bit d15 on the left, d0 on the right) to the serial interface. this writes the mid-scale code (100000000000) to latch a and also copies the full-scale code from the double buffer to latch b. both dac outputs thus begin to rise after the second write.
WM2638 production data wolfson microelectronics ltd pd rev 1.0 july 99 11 package dimensions dm009.b d: 8 pin soic 3.9 mm wide body symbols dimensions (mm) dimensions (inches) min max min max a 1.35 1.75 0.0532 0.0688 a 1 0.10 0.25 0.0040 0.0098 b 0.33 0.51 0.0130 0.0200 c 0.19 0.25 0.0075 0.0098 d 4.80 5.00 0.1890 0.1968 e 1.27 bsc 0.050 bsc e 3.80 4.00 0.1497 0.1574 h 0.25 0.50 0.0099 0.0196 h 5.80 6.20 0.2284 0.2440 l 0.40 1.27 0.0160 0.0500 a a 0 o 8 o 0 o 8 o ref: jedec.95, ms-012 notes: a. all linear dimensions are in millimeters (inches). b. this drawing is subject to change without notice. c. body dimensions do not include mold flash or protrusion, not to exceed 0.25mm (0.010in). d. meets jedec.95 ms-012, variation = aa. refer to this specification for further details. c h x 45 o a l a a1 seating plane -c- 0.10 (0.004) 4 1 d 5 8 e h b e

▲Up To Search▲

Price & Availability of WM2638

	To Download WM2638 Datasheet File
If you can't view the Datasheet, Please click here to try to view without PDF Reader .