DAC0830/DAC0831/DAC0832 8-Bit mP Compatible, Double-Buffered D to A Converters General Description Features The DAC0830 is an advanced CMOS/Si-Cr 8-bit multiplying DAC designed to interface directly with the 8080, 8048, 8085, Z80É, and other popular microprocessors. A deposited silicon-chromium R-2R resistor ladder network divides the reference current and provides the circuit with excellent temperature tracking characteristics (0.05% of Full Scale Range maximum linearity error over temperature). The circuit uses CMOS current switches and control logic to achieve low power consumption and low output leakage current errors. Special circuitry provides TTL logic input voltage level compatibility. Double buffering allows these DACs to output a voltage corresponding to one digital word while holding the next digital word. This permits the simultaneous updating of any number of DACs. The DAC0830 series are the 8-bit members of a family of microprocessor-compatible DACs (MICRO-DACTM ). For applications demanding higher resolution, the DAC1000 series (10-bits) and the DAC1208 and DAC1230 (12-bits) are available alternatives. Y Y Y Y Y Y Y Y Y Key Specifications Y Y Y Y Y BI-FETTM and MICRO-DACTM are trademarks of National Semiconductor Corporation. Z80É is a registered trademark of Zilog Corporation. Double-buffered, single-buffered or flow-through digital data inputs Easy interchange and pin-compatible with 12-bit DAC1230 series Direct interface to all popular microprocessors Linearity specified with zero and full scale adjust onlyÐ NOT BEST STRAIGHT LINE FIT. Works with g 10V reference-full 4-quadrant multiplication Can be used in the voltage switching mode Logic inputs which meet TTL voltage level specs (1.4V logic threshold) Operates ‘‘STAND ALONE’’ (without mP) if desired Available in 20-pin small-outline or molded chip carrier package Y Current settling time Resolution Linearity (guaranteed over temp.) Gain Tempco Low power dissipation Single power supply 1 ms 8 bits 8, 9, or 10 bits 0.0002% FS/§ C 20 mW 5 to 15 VDC Typical Application *Allows easy upgrade to 12-bit DAC1230, See application hints TL/H/5608 – 1 Connection Diagrams (Top Views) Dual-In-Line and Small-Outline Packages Molded Chip Carrier Package ² This is necessary for the 12-bit DAC1230 series to permit interchanging from an 8-bit to a 12-bit DAC with No PC board changes and no software changes, See applications section. TL/H/5608 – 22 TL/H/5608 – 21 C1995 National Semiconductor Corporation TL/H/5608 RRD-B30M115/Printed in U. S. A. DAC0830/DAC0831/DAC0832 8-Bit mP Compatible, Double-Buffered D to A Converters February 1995 Absolute Maximum Ratings (Notes 1 & 2) Lead Temperature (soldering, 10 sec.) Dual-In-Line Package (plastic) Dual-In-Line Package (ceramic) Surface Mount Package Vapor Phase (60 sec.) Infrared (15 sec.) If Military/Aerospace specified devices are required, please contact the National Semiconductor Sales Office/Distributors for availability and specifications. Supply Voltage (VCC) Voltage at Any Digital Input Voltage at VREF Input Storage Temperature Range Package Dissipation at TA e 25§ C (Note 3) DC Voltage Applied to IOUT1 or IOUT2 (Note 4) ESD Susceptability (Note 14) 17 VDC VCC to GND g 25V b 65§ C to a 150§ C 260§ C 300§ C 215§ C 220§ C Operating Conditions Temperature Range Part numbers with ‘LCN’ suffix Part numbers with ‘LCWM’ suffix Part numbers with ‘LCV’ suffix Part numbers with ‘LCJ’ suffix Part numbers with ‘LJ’ suffix 500 mW b 100 mV to VCC 800V TMINsTAsTMAX 0§ C to a 70§ C 0§ C to a 70§ C 0§ C to a 70§ C b 40§ C to a 85§ C b 55§ C to a 125§ C VCC to GND Voltage at Any Digital Input Electrical Characteristics VREF e 10.000 VDC unless otherwise noted. Boldface limits apply over temperature, TMINsTAsTMAX. For all other limits TA e 25§ C. Parameter VCC e 5 VDC g 5% VCC e 4.75 VDC VCC e 12 VDC g 5% VCC e 15.75 VDC Limit See to 15 VDC g 5% Note Units Tested Typ Design Limit Limit (Note 12) (Note 6) (Note 5) Conditions CONVERTER CHARACTERISTICS Resolution 8 Linearity Error Max Zero and full scale adjusted b 10V s VREF s a 10V DAC0830LJ & LCJ DAC0832LJ & LCJ DAC0830LCN, LCWM & LCV DAC0831LCN DAC0832LCN, LCWM & LCV Differential Nonlinearity Max DAC0830LJ & LCJ DAC0832LJ & LCJ DAC0830LCN, LCWM & LCV DAC0831LCN DAC0832LCN, LCWM & LCV Zero and full scale adjusted b 10V s VREF s a 10V Monotonicity b 10V s VREF s a 10V Gain Error Max Using Internal Rfb b 10V s VREF s a 10V Gain Error Tempco Max Using internal Rfb Power Supply Rejection All digital inputs latched high VCC e 14.5V to 15.5V 11.5V to 12.5V 4.5V to 5.5V Reference Input 8 8 bits 0.05 0.2 0.05 0.1 0.2 0.05 0.2 0.05 0.1 0.2 % FSR % FSR % FSR % FSR % FSR 0.1 0.4 0.1 0.2 0.4 0.1 0.4 0.1 0.2 0.4 % FSR % FSR % FSR % FSR % FSR 8 8 8 8 bits bits g1 g1 % FS 0.0006 % FS/§ C 4, 8 4, 8 LJ & LCJ LCN, LCWM & LCV 4 7 g 0.2 0.0002 0.0002 0.0006 0.013 0.0025 % FSR/V 0.015 Max 15 20 20 kX Min 15 10 10 kX Output Feedthrough Error VREF e 20 Vp-p, f e 100 kHz All data inputs latched low 3 2 mVp-p Electrical Characteristics VREF e 10.000 VDC unless otherwise noted. Boldface limits apply over temperature, TMINsTAsTMAX. For all other limits TA e 25§ C. (Continued) Parameter See Note Conditions VCC e 4.75 VDC VCC e 15.75 VDC Typ (Note 12) VCC e 5 VDC g 5% VCC e 12 VDC g 5% to 15 VDC g 5% Limit Units Tested Limit (Note 5) Design Limit (Note 6) 100 50 100 100 nA 100 50 100 100 nA CONVERTER CHARACTERISTICS (Continued) Output Leakage Current Max Output Capacitance IOUT1 All data inputs latched low LJ & LCJ LCN, LCWM & LCV 10 IOUT2 All data inputs latched high LJ & LCJ LCN, LCWM & LCV IOUT1 IOUT2 All data inputs latched low 45 115 pF IOUT1 IOUT2 All data inputs latched high 130 30 pF DIGITAL AND DC CHARACTERISTICS Digital Input Voltages Digital Input Currents Supply Current Drain Max Logic Low LJ 4.75V LJ 15.75V LCJ 4.75V LCJ 15.75V LCN, LCWM, LCV 0.6 0.8 0.7 0.8 0.95 0.8 LJ & LCJ LCN, LCWM, LCV 2.0 1.9 2.0 2.0 VDC VDC Min Logic High Max Digital inputs k0.8V LJ & LCJ LCN, LCWM, LCV b 50 b 200 b 160 b 200 b 200 mA mA Digital inputsl2.0V LJ & LCJ LCN, LCWM, LCV 0.1 a 10 a8 a 10 a 10 mA 1.2 3.5 1.7 3.5 2.0 Max LJ & LCJ LCN, LCWM, LCV 3 mA Electrical Characteristics VREF e 10.000 VDC unless otherwise noted. Boldface limits apply over temperature, TMINsTAsTMAX. For all other limits TA e 25§ C. (Continued) VCC e 15.75 VDC Symbol Parameter Conditions See Note Typ (Note 12) VCC e 12 VDC g 5% to 15 VDC g 5% Tested Limit (Note 5) Design Limit (Note 6) VCC e 4.75 VDC Typ (Note 12) Tested Limit (Note 5) VCC e 5 VDC g 5% Design Limit (Note 6) Limit Units AC CHARACTERISTICS ts Current Setting Time VIL e 0V, VIH e 5V tW Write and XFER Pulse Width Min VIL e 0V, VIH e 5V 11 9 100 Data Setup Time Min VIL e 0V, VIH e 5V 100 tDH Data Hold Time Min VIL e 0V, VIH e 5V tCS Control Setup Time VIL e 0V, VIH e 5V Min tDS tCH Control Hold Time VIL e 0V, VIH e 5V Min 1.0 9 9 250 320 320 375 250 320 320 375 30 30 9 9 1.0 110 250 320 0 0 0 ms 600 900 900 600 900 900 50 50 600 320 10 0 900 1100 ns 1100 0 0 Note 1: Absolute Maximum Ratings indicate limits beyond which damage to the device may occur. DC and AC electrical specifications do not apply when operating the device beyond its specified operating conditions. Note 2: All voltages are measured with respect to GND, unless otherwise specified. Note 3: The maximum power dissipation must be derated at elevated temperatures and is dictated by TJMAX, iJA, and the ambient temperature, TA. The maximum allowable power dissipation at any temperature is PD e (TJMAX b TA)/iJA or the number given in the Absolute Maximum Ratings, whichever is lower. For this device, TJMAX e 125§ C (plastic) or 150§ C (ceramic), and the typical junction-to-ambient thermal resistance of the J package when board mounted is 80§ C/W. For the N package, this number increases to 100§ C/W and for the V package this number is 120§ C/W. Note 4: For current switching applications, both IOUT1 and IOUT2 must go to ground or the ‘‘Virtual Ground’’ of an operational amplifier. The linearity error is degraded by approximately VOS d VREF. For example, if VREF e 10V then a 1 mV offset, VOS, on IOUT1 or IOUT2 will introduce an additional 0.01% linearity error. Note 5: Tested limits are guaranteed to National’s AOQL (Average Outgoing Quality Level). Note 6: Guaranteed, but not 100% production tested. These limits are not used to calculate outgoing quality levels. Note 7: Guaranteed at VREF e g 10 VDC and VREF e g 1 VDC. Note 8: The unit ‘‘FSR’’ stands for ‘‘Full Scale Range.’’ ‘‘Linearity Error’’ and ‘‘Power Supply Rejection’’ specs are based on this unit to eliminate dependence on a particular VREF value and to indicate the true performance of the part. The ‘‘Linearity Error’’ specification of the DAC0830 is ‘‘0.05% of FSR (MAX)’’. This guarantees that after performing a zero and full scale adjustment (see Sections 2.5 and 2.6), the plot of the 256 analog voltage outputs will each be within 0.05% c VREF of a straight line which passes through zero and full scale. Note 9: Boldface tested limits apply to the LJ and LCJ suffix parts only. Note 10: A 100nA leakage current with Rfb e 20k and VREF e 10V corresponds to a zero error of (100 c 10b9c 20 c 103) c 100/10 which is 0.02% of FS. Note 11: The entire write pulse must occur within the valid data interval for the specified tW, tDS, tDH, and tS to apply. Note 12: Typicals are at 25§ C and represent most likely parametric norm. Note 13: Human body model, 100 pF discharged through a 1.5 kX resistor. 4 Switching Waveform TL/H/5608 – 2 Definition of Package Pinouts Control Signals (All control signals level actuated) CS: Chip Select (active low). The CS in combination with ILE will enable WR1. ILE: Input Latch Enable (active high). The ILE in combination with CS enables WR1. WR1: Write 1. The active low WR1 is used to load the digital input data bits (DI) into the input latch. The data in the input latch is latched when WR1 is high. To update the input latchÐCS and WR1 must be low while ILE is high. WR2: Write 2 (active low). This signal, in combination with XFER, causes the 8-bit data which is available in the input latch to transfer to the DAC register. XFER: Transfer control signal (active low). The XFER will enable WR2. VREF: VCC: GND: Other Pin Functions DI0-DI7: Digital Inputs. DI0 is the least significant bit (LSB) and DI7 is the most significant bit (MSB). IOUT1: DAC Current Output 1. IOUT1 is a maximum for a digital code of all 1’s in the DAC register, and is zero for all 0’s in DAC register. IOUT2: DAC Current Output 2. IOUT2 is a constant minus IOUT1, or IOUT1 a IOUT2 e constant (I full scale for a fixed reference voltage). Rfb: Feedback Resistor. The feedback resistor is provided on the IC chip for use as the shunt feedback resistor for the external op amp which is used to provide an output voltage for the DAC. This on-chip resistor should always be used (not an external resistor) since it matches the resistors which are used in the on-chip R-2R ladder and tracks these resistors over temperature. Reference Voltage Input. This input connects an external precision voltage source to the internal R2R ladder. VREF can be selected over the range of a 10 to b 10V. This is also the analog voltage input for a 4-quadrant multiplying DAC application. Digital Supply Voltage. This is the power supply pin for the part. VCC can be from a 5 to a 15VDC. Operation is optimum for a 15VDC. The pin 10 voltage must be at the same ground potential as IOUT1 and IOUT2 for current switching applications. Any difference of potential (VOS pin 10) will result in a linearity change of VOS pin 10 3VREF For example, if VREF e 10V and pin 10 is 9mV offset from IOUT1 and IOUT2 the linearity change will be 0.03%. Pin 3 can be offset g 100mV with no linearity change, but the logic input threshold will shift. 5 Linearity Error TL/H/5608 – 3 a) End point test after zero and fs adj. b) Best straight line c) Shifting fs adj. to pass best straight line test Definition of Terms Settling Time: Settling time is the time required from a code transition until the DAC output reaches within g (/2LSB of the final output value. Full-scale settling time requires a zero to full-scale or full-scale to zero output change. Full-Scale Error: Full scale error is a measure of the output error between an ideal DAC and the actual device output. Ideally, for the DAC0830 series, full-scale is VREF b1LSB. For VREF e 10V and unipolar operation, VFULL-SCALE e 10.0000Vb39mV e 9.961V. Full-scale error is adjustable to zero. Differential Nonlinearity: The difference between any two consecutive codes in the transfer curve from the theoretical 1 LSB is differential nonlinearity. Monotonic: If the output of a DAC increases for increasing digital input code, then the DAC is monotonic. An 8-bit DAC which is monotonic to 8 bits simply means that increasing digital input codes will produce an increasing analog output. Resolution: Resolution is directly related to the number of switches or bits within the DAC. For example, the DAC0830 has 28 or 256 steps and therefore has 8-bit resolution. Linearity Error: Linearity Error is the maximum deviation from a straight line passing through the endpoints of the DAC transfer characteristic . It is measured after adjusting for zero and full-scale. Linearity error is a parameter intrinsic to the device and cannot be externally adjusted. National’s linearity ‘‘end point test’’ (a) and the ‘‘best straight line’’ test (b,c) used by other suppliers are illustrated above. The ‘‘end point test’’ greatly simplifies the adjustment procedure by eliminating the need for multiple iterations of checking the linearity and then adjusting full scale until the linearity is met. The ‘‘end point test’’ guarantees that linearity is met after a single full scale adjust. (One adjustment vs. multiple iterations of the adjustment.) The ‘‘end point test’’ uses a standard zero and F.S. adjustment procedure and is a much more stringent test for DAC linearity. Power Supply Sensitivity: Power supply sensitivity is a measure of the effect of power supply changes on the DAC full-scale output. TL/H/5608 – 4 FIGURE 1. DAC0830 Functional Diagram 6 Typical Performance Characteristics Digital Input Threshold vs. Temperature Digital Input Threshold vs. VCC Gain and Linearity Error Variation vs. Temperature Gain and Linearity Error Variation vs. Supply Voltage Write Pulse Width Data Hold Time TL/H/5608 – 5 DAC0830 Series Application Hints system to be updated to their new analog output levels simultaneously via a common strobe signal. The timing requirements and logic level convention of the register control signals have been designed to minimize or eliminate external interfacing logic when applied to most popular microprocessors and development systems. It is easy to think of these converters as 8-bit ‘‘write-only’’ memory locations that provide an analog output quantity. All inputs to these DAC’s meet TTL voltage level specs and can also be driven directly with high voltage CMOS logic in nonmicroprocessor based systems. To prevent damage to the chip from static discharge, all unused digital inputs should be tied to VCC or ground. If any of the digital inputs are inadvertantly left floating, the DAC interprets the pin as a logic ‘‘1’’. These DAC’s are the industry’s first microprocessor compatible, double-buffered 8-bit multiplying D to A converters. Double-buffering allows the utmost application flexibility from a digital control point of view. This 20-pin device is also pin for pin compatible (with one exception) with the DAC1230, a 12-bit MICRO-DAC. In the event that a system’s analog output resolution and accuracy must be upgraded, substituting the DAC1230 can be easily accomplished. By tying address bit A0 to the ILE pin, a two-byte mP write instruction (double precision) which automatically increments the address for the second byte write (starting with A0 e ‘‘1’’) can be used. This allows either an 8-bit or the 12-bit part to be used with no hardware or software changes. For the simplest 8-bit application, this pin should be tied to VCC (also see other uses in section 1.1). Analog signal control versatility is provided by a precision R2R ladder network which allows full 4-quadrant multiplication of a wide range bipolar reference voltage by an applied digital word. 1.1 Double-Buffered Operation Updating the analog output of these DAC’s in a double-buffered manner is basically a two step or double write operation. In a microprocessor system two unique system addresses must be decoded, one for the input latch controlled by the CS pin and a second for the DAC latch which is controlled by the XFER line. If more than one DAC is being driven, Figure 2 , the CS line of each DAC would typically be decoded individually, but all of the converters could share a common XFER address to allow simultaneous updating of any number of DAC’s. The timing for this operation is shown, Figure 3 . It is important to note that the analog outputs that will change after a simultaneous transfer are those from the DAC’s whose input register had been modified prior to the XFER command. 1.0 DIGITAL CONSIDERATIONS A most unique characteristic of these DAC’s is that the 8-bit digital input byte is double-buffered. This means that the data must transfer through two independently controlled 8bit latching registers before being applied to the R-2R ladder network to change the analog output. The addition of a second register allows two useful control features. First, any DAC in a system can simultaneously hold the current DAC data in one register (DAC register) and the next data word in the second register (input register) to allow fast updating of the DAC output on demand. Second, and probably more important, double-buffering allows any number of DAC’s in a 7 DAC0830 Series Application Hints (Continued) *TIE TO LOGIC 1 IF NOT NEEDED (SEE SEC. 1.1). FIGURE 2. Controlling Mutiple DACs TL/H/5608 – 6 FIGURE 3 one controlling the DAC’s to take over control of the data bus and control lines. If this second system were to use the same addresses as those decoded for DAC control (but for a different purpose) the ILE function would prevent the DAC’s from being erroneously altered. In a ‘‘Stand-Alone’’ system the control signals are generated by discrete logic. In this case double-buffering can be controlled by simply taking CS and XFER to a logic ‘‘0’’, ILE to a logic ‘‘1’’ and pulling WR1 low to load data to the input latch. Pulling WR2 low will then update the analog output. A logic ‘‘1’’ on either of these lines will prevent the changing of the analog output. The ILE pin is an active high chip select which can be decoded from the address bus as a qualifier for the normal CS signal generated during a write operation. This can be used to provide a higher degree of decoding unique control signals for a particular DAC, and thereby create a more efficient addressing scheme. Another useful application of the ILE pin of each DAC in a multiple DAC system is to tie these inputs together and use this as a control line that can effectively ‘‘freeze’’ the outputs of all the DAC’s at their present value. Pulling this line low latches the input register and prevents new data from being written to the DAC. This can be particularly useful in multiprocessing systems to allow a processor other than the 8 DAC0830 Series Application Hints (Continued) TL/H/5608 – 7 ILE e LOGIC ‘‘1’’; WR2 and XFER GROUNDED FIGURE 4 1.2 Single-Buffered Operation In a microprocessor controlled system where maximum data throughput to the DAC is of primary concern, or when only one DAC of several needs to be updated at a time, a single-buffered configuration can be used. One of the two internal registers allows the data to flow through and the other register will serve as the data latch. Digital signal feedthrough (see Section 1.5) is minimized if the input register is used as the data latch. Timing for this mode is shown in Figure 4 . Single-buffering in a ‘‘stand-alone’’ system is achieved by strobing WR1 low to update the DAC with CS, WR2 and XFER grounded and ILE tied high. be met or erroneous data can be latched. This hold time is defined as the length of time data must be held valid on the digital inputs after a qualified (via CS) WR strobe makes a low to high transition to latch the applied data. If the controlling device or system does not inherently meet these timing specs the DAC can be treated as a slow memory or peripheral and utilize a technique to extend the write strobe. A simple extension of the write time, by adding a wait state, can simultaneously hold the write strobe active and data valid on the bus to satisfy the minimum WR pulsewidth. If this does not provide a sufficient data hold time at the end of the write cycle, a negative edge triggered oneshot can be included between the system write strobe and the WR pin of the DAC. This is illustrated in Figure 5 for an exemplary system which provides a 250ns WR strobe time with a data hold time of less than 10ns. The proper data set-up time prior to the latching edge (LO to HI transition) of the WR strobe, is insured if the WR pulsewidth is within spec and the data is valid on the bus for the duration of the DAC WR strobe. 1.3 Flow-Through Operation Though primarily designed to provide microprocessor interface compatibility, the MICRO-DAC’s can easily be configured to allow the analog output to continuously reflect the state of an applied digital input. This is most useful in applications where the DAC is used in a continuous feedback control loop and is driven by a binary up-down counter, or in function generation circuits where a ROM is continuously providing DAC data. Simply grounding CS, WR1, WR2, and XFER and tying ILE high allows both internal registers to follow the applied digital inputs (flow-through) and directly affect the DAC analog output. 1.5 Digital Signal Feedthrough When data is latched in the internal registers, but the digital inputs are changing state, a narrow spike of current may flow out of the current output terminals. This spike is caused by the rapid switching of internal logic gates that are responding to the input changes. There are several recommendations to minimize this effect. When latching data in the DAC, always use the input register as the latch. Second, reducing the VCC supply for the DAC from a 15V to a 5V offers a factor of 5 improvement in the magnitude of the feedthrough, but at the expense of internal logic switching speed. Finally, increasing CC (Figure 8 ) to a value consistent with the actual circuit bandwidth requirements can provide a substantial damping effect on any output spikes. 1.4 Control Signal Timing When interfacing these MICRO-DAC to any microprocessor, there are two important time relationships that must be considered to insure proper operation. The first is the minimum WR strobe pulse width which is specified as 900 ns for all valid operating conditions of supply voltage and ambient temperature, but typically a pulse width of only 180ns is adequate if VCC e 15VDC. A second consideration is that the guaranteed minimum data hold time of 50ns should 9 DAC0830 Series Application Hints (Continued) TL/H/5608 – 8 FIGURE 5. Accommodating a High Speed System 2.0 ANALOG CONSIDERATIONS The fundamental purpose of any D to A converter is to provide an accurate analog output quantity which is representative of the applied digital word. In the case of the DAC0830, the output, IOUT1, is a current directly proportional to the product of the applied reference voltage and the digital input word. For application versatility, a second output, IOUT2, is provided as a current directly proportional to the complement of the digital input. Basically: Figure 6 . The MOS switches operate in the current mode with a small voltage drop across them and can therefore switch currents of either polarity. This is the basis for the 4quadrant multiplying feature of this DAC. 2.2 Basic Unipolar Output Voltage To maintain linearity of output current with changes in the applied digital code, it is important that the voltages at both of the current output pins be as near ground potential (0VDC) as possible. With VREF e a 10V every millivolt appearing at either IOUT1 or IOUT2 will cause a 0.01% linearity error. In most applications this output current is converted to a voltage by using an op amp as shown in Figure 7 . The inverting input of the op amp is a ‘‘virtual ground’’ created by the feedback from its output through the internal 15 kX resistor, Rfb. All of the output current (determined by the digital input and the reference voltage) will flow through Rfb to the output of the amplifier. Two-quadrant operation can be obtained by reversing the polarity of VREF thus causing IOUT1 to flow into the DAC and be sourced from the output of the amplifier. The output voltage, in either case, is always equal to IOUT1 c Rfb and is the opposite polarity of the reference voltage. The reference can be either a stable DC voltage source or an AC signal anywhere in the range from b10V to a 10V. The DAC can be thought of as a digitally controlled attenuator: the output voltage is always less than or equal to the applied reference voltage. The VREF terminal of the device presents a nominal impedance of 15 kX to ground to external circuitry. Always use the internal Rfb resistor to create an output voltage since this resistor matches (and tracks with temperature) the value of the resistors used to generate the output current (IOUT1). VREF Digital Input c ; 15 kX 256 VREF 255bDigital Input c IOUT2 e 15 kX 256 where the digital input is the decimal (base 10) equivalent of the applied 8-bit binary word (0 to 255), VREF is the voltage at pin 8 and 15 kX is the nominal value of the internal resistance, R, of the R-2R ladder network (discussed in Section 2.1). Several factors external to the DAC itself must be considered to maintain analog accuracy and are covered in subsequent sections. IOUT1 e 2.1 The Current Switching R-2R Ladder The analog circuitry, Figure 6 , consists of a silicon-chromium (SiCr or Si-chrome) thin film R-2R ladder which is deposited on the surface oxide of the monolithic chip. As a result, there are no parasitic diode problems with the ladder (as there may be with diffused resistors) so the reference voltage, VREF, can range b10V to a 10V even if VCC for the device is 5VDC. The digital input code to the DAC simply controls the position of the SPDT current switches and steers the available ladder current to either IOUT1 or IOUT2 as determined by the logic input level (‘‘1’’ or ‘‘0’’) respectively, as shown in 10 DAC0830 Series Application Hints (Continued) FIGURE 6 FIGURE 7 TL/H/5608 – 9 2.3 Op Amp Considerations This configuration features several improvements over exThe op amp used in Figure 7 should have offset voltage isting circuits for bipolar outputs with other multiplying nulling capability (See Section 2.5). DACs. Only the offset voltage of amplifier 1 has to be nulled The selected op amp should have as low a value of input to preserve linearity of the DAC. The offset voltage error of bias current as possible. The product of the bias current the second op amp (although a constant output voltage ertimes the feedback resistance creates an output voltage error) has no effect on linearity. It should be nulled only if ror which can be significant in low reference voltage appliabsolute output accuracy is required. Finally, the values of cations. BI-FET op amps are highly recommended for use the resistors around the second amplifier do not have to with these DACs because of their very low input current. match the internal DAC resistors, they need only to match Transient response and settling time of the op amp are imand temperature track each other. A thin film 4-resistor netportant in fast data throughput applications. The largest stawork available from Beckman Instruments, Inc. (part no. bility problem is the feedback pole created by the feedback 694-3-R10K-D) is ideally suited for this application. These resistance, Rfb, and the output capacitance of the DAC. resistors are matched to 0.1% and exhibit only 5 ppm/§ C This appears from the op amp output to the (b) input and resistance tracking temperature coefficient. Two of the four includes the stray capacitance at this node. Addition of a available 10 kX resistors can be paralleled to form R in lead capacitance, CC in Figure 8 , greatly reduces overshoot Figure 9 and the other two can be used independently as and ringing at the output for a step change in DAC output the resistances labeled 2R. current. 2.5 Zero Adjustment Finally, the output voltage swing of the amplifier must be greater than VREF to allow reaching the full scale output For accurate conversions, the input offset voltage of the voltage. Depending on the loading on the output of the amoutput amplifier must always be nulled. Amplifier offset erplifier and the available op amp supply voltages (only g 12 rors create an overall degradation of DAC linearity. volts in many development systems), a reference voltage The fundamental purpose of zeroing is to make the voltage less than 10 volts may be necessary to obtain the full anaappearing at the DAC outputs as near 0VDC as possible. log output voltage range. This is accomplished for the typical DAC Ð op amp connection (Figure 7 ) by shorting out Rfb, the amplifier feedback 2.4 Bipolar Output Voltage with a Fixed Reference resistor, and adjusting the VOS nulling potentiometer of the The addition of a second op amp to the previous circuitry op amp until the output reads zero volts. This is done, of can be used to generate a bipolar output voltage from a course, with an applied digital code of all zeros if IOUT1 is fixed reference voltage. This, in effect, gives sign signifidriving the op amp (all one’s for IOUT2). The short around cance to the MSB of the digital input word and allows twoRfb is then removed and the converter is zero adjusted. quadrant multiplication of the reference voltage. The polarity of the reference can also be reversed to realize full 4-quadrant multiplication: g VREF c g Digital Code e g VOUT. This circuit is shown in Figure 9 . 11 DAC0830 Series Application Hints (Continued) OP Amp CC ts (O to Full Scale) LF356 LF351 LF357* 22 pF 22 pF 10 pF 4 ms 5 ms 2 ms *2.4 kX RESISTOR ADDED FROM b INPUT TO GROUND TO INSURE STABILITY FIGURE 8 VOUT e VREF 1 LSB e lVREFl TL/H/5608 – 10 128 Input Code IDEAL VOUT MSB ÀÀÀÀÀÀÀÀÀÀLSB *THESE RESISTORS ARE AVAILABLE FROM BECKMAN INSTRUMENTS, INC. AS THEIR PART NO. 694-3-R10K-D 1 1 1 0 1 1 0 1 (DIGITAL CODEb128) 128 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0 1 a VREF b VREF VREFb1 LSB b VREF a 1 LSB b VREF /2 VREF/2 0 b 1 LSB lVREFl b1 LSB 0 0 1 1 1 1 1 1 b 2 b l VREF l 0 0 0 0 0 0 0 0 l l l l 0 a 1 LSB lVREFl a 1 LSB 2 l a VREF l FIGURE 9 2.6 Full-Scale Adjustment manner from the standard current switching configuration. In the case where the matching of Rfb to the R value of the The reference voltage is connected to one of the current R-2R ladder (typically g 0.2%) is insufficient for full-scale output terminals (IOUT1 for true binary digital control, IOUT2 accuracy in a particular application, the VREF voltage can be adjusted or an external resistor and potentiometer can be is for complementary binary) and the output voltage is taken added as shown in Figure 10 to provide a full-scale adjustfrom the normal VREF pin. The converter output is now a ment. voltage in the range from 0V to 255/256 VREF as a function of the applied digital code as shown in Figure 11 . The temperature coefficients of the resistors used for this adjustment are an important concern. To prevent degradation of the gain error temperature coefficient by the external resistors, their temperature coefficients ideally would have to match that of the internal DAC resistors, which is a highly impractical constraint. For the values shown in Figure 10 , if the resistor and the potentiometer each had a temperature coefficient of g 100 ppm/§ C maximum, the overall gain error temperature coefficent would be degraded a maximum of 0.0025%/§ C for an adjustment pot setting of less than 3% of Rfb. 2.7 Using the DAC0830 in a Voltage Switching Configuration The R-2R ladder can also be operated as a voltage switching network. In this mode the ladder is used in an inverted TL/H/5608 – 11 FIGURE 10. Adding Full-Scale Adjustment 12 DAC0830 Series Application Hints (Continued) TL/H/5608 – 12 FIGURE 11. Voltage Mode Switching gain error on the voltage difference between VCC and the voltage applied to the normal current output terminals. This is a result of the voltage drive requirements of the ladder switches. To ensure that all 8 switches turn on sufficiently (so as not to add significant resistance to any leg of the ladder and thereby introduce additional linearity and gain errors) it is recommended that the applied reference voltage be kept less than a 5VDC and VCC be at least 9V more positive than VREF. These restrictions ensure less than 0.1% linearity and gain error change. Figures 16, 17 and 18 characterize the effects of bringing VREF and VCC closer together as well as typical temperature performance of this voltage switching configuration. This configuration offers several useful application advantages. Since the output is a voltage, an external op amp is not necessarily required but the output impedance of the DAC is fairly high (equal to the specified reference input resistance of 10 kX to 20 kX) so an op amp may be used for buffering purposes. Some of the advantages of this mode are illustrated in Figures 12, 13, 14 and 15 . There are two important things to keep in mind when using this DAC in the voltage switching mode. The applied reference voltage must be positive since there are internal parasitic diodes from ground to the IOUT1 and IOUT2 terminals which would turn on if the applied reference went negative. There is also a dependence of conversion linearity and TL/H/5608 – 13 # Voltage switching mode eliminates output signal inversion and therefore a # VOUT e 2.5V need for a negative power supply. # Zero code output voltage is limited by the low level output saturation volt- # D b1 128 J # Slewing and settling time for a full scale output change is & 1.8 ms age of the op amp. The 2 kX pull-down resistor helps to reduce this voltage. # VOS of the op amp has no effect on DAC linearity. FIGURE 13. Obtaining a Bipolar Output from a Fixed Reference with a Single Op Amp FIGURE 12. Single Supply DAC 13 DAC0830 Series Application Hints (Continued) FIGURE 14. Bipolar Output with Increased Output Voltage Swing TL/H/5608 – 14 # Only a single a 15V supply required # Non-interactive full-scale and zero code output adjustments # VMAX and VMIN must be s a 5VDC and t 0V. # Incremental Output Step e # VOUT e 1 (VMAX b VMIN). 256 D 255 (VMAX b VMIN) a VMIN 256 256 FIGURE 15. Single Supply DAC with Level Shift and SpanAdjustable Output Gain and Linearity Error Variation vs. Supply Voltage Gain and Linearity Error Variation vs. Reference Voltage Gain and Linearity Error Variation vs. Temperature TL/H/5608 – 15 FIGURE 16 FIGURE 17 Note: For these curves, VREF is the voltage applied to pin 11 (IOUT1) with pin 12 (IOUT2) grounded. 14 FIGURE 18 DAC0830 Series Application Hints (Continued) 2.8 Miscellaneous Application Hints These converters are CMOS products and reasonable care should be exercised in handling them to prevent catastrophic failures due to static discharge. Conversion accuracy is only as good as the applied reference voltage so providing a stable source over time and temperature changes is an important factor to consider. A ‘‘good’’ ground is most desirable. A single point ground distribution technique for analog signals and supply returns keeps other devices in a system from affecting the output of the DACs. During power-up supply voltage sequencing, the b15V (or b 12V) supply of the op amp may appear first. This will cause the output of the op amp to bias near the negative supply potential. No harm is done to the DAC, however, as the on-chip 15 kX feedback resistor sufficiently limits the current flow from IOUT1 when this lead is internally clamped to one diode drop below ground. Careful circuit construction with minimization of lead lengths around the analog circuitry, is a primary concern. Good high frequency supply decoupling will aid in preventing inadvertant noise from appearing on the analog output. Overall noise reduction and reference stability is of particular concern when using the higher accuracy versions, the DAC0830 and DAC0831, or their advantages are wasted. 3.0 GENERAL APPLICATION IDEAS The connections for the control pins of the digital input registers are purposely omitted. Any of the control formats discussed in Section 1 of the accompanying text will work with any of the circuits shown. The method used depends on the overall system provisions and requirements. The digital input code is referred to as D and represents the decimal equivalent value of the 8-bit binary input, for example: Binary Input Pin 13 MSB 1 1 0 0 0 1 0 0 0 0 Pin 7 LSB 1 0 0 0 0 1 0 1 0 0 1 0 0 0 0 1 0 0 0 0 1 0 0 1 0 D Decimal Equivalent 1 0 0 0 0 255 128 16 2 0 Applications DAC Controlled Amplifier (Volume Control) Capacitance Multiplier TL/H/5608 – 16 # VOUT e # b VIN (256) # CEQUIV e C1 1 a D # When D e 0, the amplifier will go open loop and the output will saturate. # Feedback impedance from the b input to the output varies from 15 kX to 256 D J # Maximum voltage across the equivalent capacitance is limited to % as the input code changes from full-scale to zero. VO MAX (op amp) 256 1a D # C2 is used to improve settling time of op amp. 15 Applications (Continued) Variable fO, Variable QO, Constant BW Bandpass Filter TL/H/5608 – 17 0 KD 256 ; QO e # fO e 2 q R 1C 0256 R (K KD (2RQ a R1) RQ(K a 1) ; 3dbBW e a 1) 2qR1C(2RQ a R1) Q where C1 e C2 e C; K e R6 and R1 e R of DAC e 15k R5 # HO e 1 for RIN e R4 e R1 # Range of fO and Q is & 16 to 1 for circuit shown. The range can be extended to 255 to 1 by replacing R1 with a second DAC0830 driven by the same digital input word. # Maximum fO c Q product should be s 200 kHz. DAC Controlled Function Generator TL/H/5608 – 18 # DAC controls the frequency of sine, square, and triangle outputs. #fe D for VOMAX e V0MIN of square wave output and R1 e 3 R2. 256(20k)C # 255 to 1 linear frequency range; oscillator stops with D e 0 # Trim symmetry and wave-shape for minimum sine wave distortion. 16 Applications (Continued) Two Terminal Floating 4 to 20 mA Current Loop Controller TL/H/5608 – 19 IOUT e VREF # DAC0830 linearly controls the current flow from the input terminal to the Ð 1 D a R1 256 Rfb ( Ð1 a R2 R3 ( output terminal to be 4 mA (for D e 0) to 19.94 mA (for D e 255). # Circuit operates with a terminal voltage differential of 16V to 55V. # P2 adjusts the magnitude of the output current and P1 adjusts the zero to full scale range of output current. # Digital inputs can be supplied from a processor using opto isolators on each input or the DAC latches can flow-through (connect control lines to pins 3 and 10 of the DAC) and the input data can be set by SPST toggle switches to ground (pins 3 and 10). DAC Controlled Exponential Time Response TL/H/5608 – 20 # Output responds exponentially to input changes and automatically stops when VOUT e VIN # Output time constant is directly proportional to the DAC input code and capacitor C # Input voltage must be positive (See section 2.7) 17 Ordering Information 0§ C to a 70§ Temperature Range Non Linearity 0.05% FSR DAC0830LCN 0.1% FSR DAC0831LCN 0.2% FSR Package Outline DAC0830LCM b 40§ C to a 85§ C b 55§ C to a 125§ C DAC0830LCV DAC0830LCJ DAC0830LJ DAC0832LCJ DAC0832LJ DAC0832LCN DAC0832LCM DAC0832LCV N20AÐMolded DIP M20B Small Outline V20A Chip Carrier Physical Dimensions inches (millimeters) Ceramic Dual-In-Line Package (J) Order Number DAC0830LCJ, DAC0830LJ, DAC0832LJ or DAC0832LCJ NS Package Number J20A 18 J20AÐCeramic DIP Physical Dimensions inches (millimeters) (Continued) Molded Small Outline Package (M) Order Number DAC0830LCM or DAC0832LCM NS Package Number M20B Molded Dual-In-Line Package (N) Order Number DAC0830LCN, DAC0831LCN or DAC0832LCN NS Package Number N20A 19 DAC0830/DAC0831/DAC0832 8-Bit mP Compatible, Double-Buffered D to A Converters Physical Dimensions inches (millimeters) (Continued) Molded Chip Carrier (V) Order Number DAC0830LCV or DAC0832LCV NS Package Number V20A LIFE SUPPORT POLICY NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein: 1. Life support devices or systems are devices or systems which, (a) are intended for surgical implant into the body, or (b) support or sustain life, and whose failure to perform, when properly used in accordance with instructions for use provided in the labeling, can be reasonably expected to result in a significant injury to the user. National Semiconductor Corporation 1111 West Bardin Road Arlington, TX 76017 Tel: 1(800) 272-9959 Fax: 1(800) 737-7018 2. A critical component is any component of a life support device or system whose failure to perform can be reasonably expected to cause the failure of the life support device or system, or to affect its safety or effectiveness. National Semiconductor Europe Fax: (a49) 0-180-530 85 86 Email: cnjwge @ tevm2.nsc.com Deutsch Tel: (a49) 0-180-530 85 85 English Tel: (a49) 0-180-532 78 32 Fran3ais Tel: (a49) 0-180-532 93 58 Italiano Tel: (a49) 0-180-534 16 80 National Semiconductor Hong Kong Ltd. 13th Floor, Straight Block, Ocean Centre, 5 Canton Rd. Tsimshatsui, Kowloon Hong Kong Tel: (852) 2737-1600 Fax: (852) 2736-9960 National Semiconductor Japan Ltd. Tel: 81-043-299-2309 Fax: 81-043-299-2408 National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.