1 ppm, 20-Bit, ±1 LSB INL,

Voltage Output DAC

Data Sheet AD5791
FEATURES FUNCTIONAL BLOCK DIAGRAM
VCC VDD VREFPF VREFPS
1 ppm resolution
1 ppm INL
AD5791 6.8kΩ 6.8kΩ
7.5 nV/√Hz noise spectral density IOVCC A1 RFB
R1 RFB
0.19 LSB long-term linearity stability INV
SDIN
<0.05 ppm/°C temperature drift INPUT
SHIFT 20
DAC
20 20-BIT
SCLK VOUT
1 µs settling time SYNC
REGISTER
AND
REG DAC
CONTROL
1.4 nV-sec glitch impulse SDO LOGIC
Operating temperature range: −40°C to +125°C 6kΩ
LDAC
20-lead TSSOP package
Wide power supply range up to ±16.5 V CLR POWER-ON-RESET
RESET AND CLEAR LOGIC
35 MHz Schmitt triggered digital interface

08964-001
1.8 V compatible digital interface DGND VSS AGND VREFNF VREFNS

APPLICATIONS Figure 1.

Medical instrumentation
Table 1. Complementary Devices
Test and measurement
Model Description
Industrial control
AD8675 Ultra precision, 36 V, 2.8 nV/√Hz rail-to-rail
High end scientific and aerospace instrumentation
output op amp
AD8676 Ultra precision, 36 V, 2.8 nV/√Hz dual rail-to-
rail output op amp
ADA4898-1 High voltage, low noise, low distortion,
unity-gain stable, high speed op amp

Table 2. Related Device

Model Description
AD5781 18-bit, 0.5 LSB INL, voltage output DAC
GENERAL DESCRIPTION
The AD57911 is a single 20-bit, unbuffered voltage-output digital- output powers up to 0 V and in a known output impedance
to-analog converter (DAC) that operates from a bipolar supply of state and remains in this state until a valid write to the device
up to 33 V. The AD5791 accepts a positive reference input in the takes place. The device provides an output clamp feature that
range 5 V to VDD − 2.5 V and a negative reference input in the places the output in a defined load state.
range VSS + 2.5 V to 0 V. The AD5791 offers a relative accuracy
PRODUCT HIGHLIGHTS
specification of ±1 LSB max, and operation is guaranteed
monotonic with a ±1 LSB differential nonlinearity (DNL) 1. 1 ppm Accuracy.
maximum specification. 2. Wide Power Supply Range up to ±16.5 V.
3. Operating Temperature Range: −40°C to +125°C.
The device uses a versatile 3-wire serial interface that operates 4. Low 7.5 nV/√Hz Noise Spectral Density.
at clock rates up to 35 MHz and that is compatible with 5. Low 0.05 ppm/°C Temperature Drift.
standard serial peripheral interface (SPI), QSPI™,
MICROWIRE™, and DSP interface standards. The device
incorporates a power-on reset circuit that ensures the DAC

1
Protected by U.S. Patent No. 7,884,747. Other patents pending.

Rev. F Document Feedback

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 One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
license is granted by implication or otherwise under any patent or patent rights of Analog Devices. Tel: 781.329.4700 ©2010–2020 Analog Devices, Inc. All rights reserved.
Trademarks and registered trademarks are the property of their respective owners. Technical Support www.analog.com
AD5791 Data Sheet

TABLE OF CONTENTS
Features .............................................................................................. 1 Serial Interface ............................................................................ 19
Applications ....................................................................................... 1 Hardware Control Pins .............................................................. 20
Functional Block Diagram .............................................................. 1 On-Chip Registers ...................................................................... 21
General Description ......................................................................... 1 AD5791 Features ............................................................................ 24
Product Highlights ........................................................................... 1 Power-On to 0 V......................................................................... 24
Revision History ............................................................................... 2 Power-Up Sequence ................................................................... 24
Specifications..................................................................................... 3 Configuring the AD5791 .......................................................... 24
Timing Characteristics ................................................................ 5 DAC Output State ...................................................................... 24
Absolute Maximum Ratings............................................................ 7 Linearity Compensation ............................................................ 24
ESD Caution .................................................................................. 7 Output Amplifier Configuration.............................................. 24
Pin Configuration and Function Descriptions ............................. 8 Applications Information .............................................................. 26
Typical Performance Characteristics ............................................. 9 Typical Operating Circuit ......................................................... 26
Terminology .................................................................................... 17 Outline Dimensions ....................................................................... 27
Theory of Operation ...................................................................... 19 Ordering Guide .......................................................................... 27
DAC Architecture ....................................................................... 19

REVISION HISTORY
1/2020—Rev. E to Rev. F 9/2011—Rev. A to Rev. B
Change to Endnote 2, Table 4 ......................................................... 5 Added Patent Note ............................................................................1
Change to Input Shift Register Section ........................................ 19 Changes to Table 3.............................................................................3
Changes to OPGND Description Column, Table 12 ................. 23
4/2018—Rev. D to Rev. E Change to Figure 51 ....................................................................... 25
Change to Figure 49 ....................................................................... 19
Added Power-Up Sequence Section and Figure 50; Renumbered 8/2011—Rev. 0 to Rev. A
Sequentially ..................................................................................... 24 Change to Features Section ..............................................................1
Changes to Specifications Section, Table 3 ....................................3
7/2013—Rev. C to Rev. D Deleted t14 Timing Specification in Table 4, Renumbered
Change to Table 4 ............................................................................. 5 Subsequent Timing Parameters Sequentially ................................5
Deleted Figure 4, Renumbered Sequentially................................. 7 Changes to Figure 2 and Figure 3 ....................................................6
Deleted Daisy-Chain Operation Section and Figure 51 ............ 21 Changes to Figure 4 ...........................................................................7
Changes to Figure 42...................................................................... 16
11/2011—Rev. B to Rev. C Changes to Figure 43...................................................................... 16
Added Figure 48; Renumbered Sequentially .............................. 17 Added Figure 44, Figure 45, and Figure 46, Renumbered
Change to Ideal Transfer Function Equation.............................. 22 Sequentially ..................................................................................... 16

7/2010—Revision 0: Initial Version

Rev. F | Page 2 of 27
Data Sheet AD5791

SPECIFICATIONS
VDD = 12.5 V to 16.5 V, VSS = −16.5 V to −12.5 V, VREFP = 10 V, VREFN = −10 V, VCC = 2.7 V to +5.5 V, IOVCC = 1.71 V to 5.5 V,
RL = unloaded, CL = unloaded, all specifications TMIN to TMAX, unless otherwise noted.

Table 3.
A, B Version 1
Parameter Min Typ Max Unit Test Conditions/Comments
STATIC PERFORMANCE 2
Resolution 20 Bits
Integral Nonlinearity Error (Relative Accuracy) −1 ±0.25 +1 LSB B version, VREFP = +10 V, VREFN = −10 V,
TA = 0°C to 105°C
−1.5 ±0.25 +1.5 LSB B version, VREFP = +10 V, VREFN = −10 V
−1.5 ±0.5 +1.5 LSB B version, VREFP = 10 V, VREFN = 0 V 3
−3 ±1 +3 LSB B version, VREFP = 5 V, VREFN = 0 V3
−4 ±2 +4 LSB A version 4
Differential Nonlinearity Error −1 ±0.5 +1 LSB VREFP = +10 V, VREFN = −10 V
−1.5 ±0.75 +1.5 LSB VREFP = 10 V, VREFN = 0 V
−2.5 ±1 +2.5 LSB VREFP = 5 V, VREFN = 0 V
Linearity Error Long Term Stability 5 0.16 LSB After 500 hours at TA = 125°C
0.19 LSB After 1000 hours at TA = 125°C
0.11 LSB After 1000 hours at TA = 100°C
Full-Scale Error −7 ±0.1 +7 LSB VREFP = +10 V, VREFN = −10 V3
−11 ±0.25 +11 LSB VREFP = 10 V, VREFN = 0 V3
−21 ±0.8 +21 LSB VREFP = 5 V, VREFN = 0 V3
−4 ±0.1 +4 LSB VREFP = +10 V, VREFN = −10 V3, TA = 0°C to 105°C
−4 ±0.25 +4 LSB VREFP = 10 V, VREFN = 0 V3, TA = 0°C to 105°C
−6 ±0.8 +6 LSB VREFP = 5 V, VREFN = 0 V3, TA = 0°C to 105°C
Full-Scale Error Temperature Coefficient ±0.02 ppm FSR/°C
Zero-Scale Error −7 ±0.1 +7 LSB VREFP = +10 V, VREFN = −10 V3
−10 ±0.15 +10 LSB VREFP = 10 V, VREFN = 0 V3
−21 ±0.75 +21 LSB VREFP = 5 V, VREFN = 0 V3
−4 ±0.1 +4 LSB VREFP = +10 V, VREFN = −10 V3, TA = 0°C to 105°C
−4 ±0.15 +4 LSB VREFP = 10 V, VREFN = 0 V3, TA = 0°C to 105°C
−6 ±0.75 +6 LSB VREFP = 5 V, VREFN = 0 V3, TA = 0°C to 105°C
Zero-Scale Error Temperature Coefficient3 ±0.04 ppm FSR/°C
Gain Error −6 ±0.3 +6 ppm FSR VREFP = +10 V, VREFN = −10 V3
−10 ±0.4 +10 ppm FSR VREFP = 10 V, VREFN = 0 V3
−20 ±0.4 +20 ppm FSR VREFP = 5 V, VREFN = 0 V3
−6 ±0.3 +6 ppm FSR VREFP = +10 V, VREFN = −10 V3, TA = 0°C to 105°C
−6 ±0.4 +6 ppm FSR VREFP = 10 V, VREFN = 0 V3, TA = 0°C to 105°C
−7 ±0.4 +7 ppm FSR VREFP = 5 V, VREFN = 0 V3, TA = 0°C to 105°C
Gain Error Temperature Coefficient3 ±0.04 ppm FSR/°C
R1, RFB Matching 0.01 %
OUTPUT CHARACTERISTICS3
Output Voltage Range VREFN VREFP V
Output Slew Rate 50 V/µs
Output Voltage Settling Time 1 µs 10 V step to 0.02%, using the AD845 buffer
in unity-gain mode
1 µs 500 code step to ±1 LSB6
Output Noise Spectral Density 7.5 nV/√Hz at 1 kHz, DAC code = midscale
7.5 nV/√Hz at 10 kHz, DAC code = midscale
7.5 nV/√Hz At 100 kHz, DAC code = midscale
Output Voltage Noise 1.1 µV p-p DAC code = midscale, 0.1 Hz to 10 Hz
bandwidth 7

Rev. F | Page 3 of 27
AD5791 Data Sheet
A, B Version 1
Parameter Min Typ Max Unit Test Conditions/Comments
Midscale Glitch Impulse 8 3.1 nV-sec VREFP = +10 V, VREFN = −10 V
1.7 nV-sec VREFP = 10 V, VREFN = 0 V
1.4 nV-sec VREFP = 5 V, VREFN = 0 V
MSB Segment Glitch Impulse8 9.1 nV-sec VREFP = +10 V, VREFN = −10 V, see Figure 42
3.6 nV-sec VREFP = 10 V, VREFN = 0 V, see Figure 43
1.9 nV-sec VREFP = 5 V, VREFN = 0 V, see Figure 44
Output Enabled Glitch Impulse 45 nV-sec On removal of output ground clamp
Digital Feedthrough 0.4 nV-sec
DC Output Impedance (Normal Mode) 3.4 kΩ
DC Output Impedance (Output Clamped 6 kΩ
to Ground)
Spurious Free Dynamic Range 100 dB 1 kHz tone, 10 kHz sample rate
Total Harmonic Distortion 97 dB 1 kHz tone, 10 kHz sample rate
REFERENCE INPUTS3
VREFP Input Range 5 VDD − 2.5 V V
VREFN Input Range VSS + 2.5 V 0
DC Input Impedance 5 6.6 kΩ VREFP, VREFN, code dependent,
typical at midscale code
Input Capacitance 15 pF VREFP, VREFN
LOGIC INPUTS3
Input Current 9 −1 +1 µA
Input Low Voltage, VIL 0.3 × IOVCC V IOVCC = 1.71 V to 5.5 V
Input High Voltage, VIH 0.7 × IOVCC V IOVCC = 1.71 V to 5.5 V
Pin Capacitance 5 pF
LOGIC OUTPUT (SDO)3
Output Low Voltage, VOL 0.4 V IOVCC = 1.71 V to 5.5 V, sinking 1 mA
Output High Voltage, VOH IOVCC − 0.5 V V IOVCC = 1.71 V to 5.5 V, sourcing 1 mA
High Impedance Leakage Current ±1 µA
High Impedance Output Capacitance 3 pF
POWER REQUIREMENTS All digital inputs at DGND or IOVCC
VDD 7.5 VSS + 33 V
VSS VDD − 33 −2.5 V
VCC 2.7 5.5 V
IOVCC 1.71 5.5 V IOVCC ≤ VCC
IDD 4.2 5.2 mA
ISS 4 4.9 mA
ICC 600 900 µA
IOICC 52 140 µA SDO disabled
DC Power Supply Rejection Ratio3, 10 ±0.6 µV/V VDD ± 10%, VSS = 15 V
±0.6 µV/V VSS ± 10%, VDD = 15 V
AC Power Supply Rejection Ratio3 95 dB VDD ± 200 mV, 50 Hz/60 Hz, VSS = −15 V
95 dB ∆VSS ± 200 mV, 50 Hz/60 Hz, VDD = 15 V
1
Temperature range: −40°C to +125°C, typical at +25°C and VDD = +15 V, VSS = −15 V, VREFP = +10 V, VREFN = −10 V.
2
Performance characterized with AD8676BRZ voltage reference buffers and AD8675ARZ output buffer.
3
Guaranteed by design and characterization, not production tested.
4
Valid for all voltage reference spans.
5
Linearity error refers to both INL error and DNL error. Either parameter can be expected to drift by the amount specified after the length of time specified.
6
AD5791 configured in X2 gain mode, 25 pF compensation capacitor on AD797.
7
Includes noise contribution from AD8676BRZ voltage reference buffers.
8
The AD5791 is configured in bias compensation mode with a low-pass RC filter on the output. R = 300 Ω, C = 143 pF (total capacitance seen by the output buffer, lead
capacitance, and so forth).
9
Current flowing in an individual logic pin.
10
Includes PSRR of AD8676BRZ voltage reference buffers.

Rev. F | Page 4 of 27
Data Sheet AD5791
TIMING CHARACTERISTICS
VCC = 2.7 V to 5.5 V; all specifications TMIN to TMAX, unless otherwise noted.

Table 4.
Limit 1
Parameter IOVCC = 1.71 V to 3.3 V IOVCC = 3.3 V to 5.5 V Unit Test Conditions/Comments
t1 2 40 28 ns min SCLK cycle time
92 60 ns min SCLK cycle time (readback mode)
t2 15 10 ns min SCLK high time
t3 9 5 ns min SCLK low time
t4 5 5 ns min SYNC to SCLK falling edge setup time
t5 2 2 ns min SCLK falling edge to SYNC rising edge hold time
t6 48 40 ns min Minimum SYNC high time
t7 8 6 ns min SYNC rising edge to next SCLK falling edge ignore
t8 9 7 ns min Data setup time
t9 12 7 ns min Data hold time
t10 13 10 ns min LDAC falling edge to SYNC falling edge
t11 20 16 ns min SYNC rising edge to LDAC falling edge
t12 14 11 ns min LDAC pulse width low
t13 130 130 ns typ LDAC falling edge to output response time
t14 130 130 ns typ SYNC rising edge to output response time (LDAC tied low)
t15 50 50 ns min CLR pulse width low
t16 140 140 ns typ CLR pulse activation time
t17 0 0 ns min SYNC falling edge to first SCLK rising edge
t18 65 60 ns max SYNC rising edge to SDO tristate (CL = 50 pF)
t19 62 45 ns max SCLK rising edge to SDO valid (CL = 50 pF)
t20 0 0 ns min SYNC rising edge to SCLK rising edge ignore
t21 35 35 ns typ RESET pulse width low
t22 150 150 ns typ RESET pulse activation time
1
All input signals are specified with tR = tF = 1 ns/V (10% to 90% of IOVCC) and timed from a voltage level of (VIL + VIH)/2.
2
Maximum SCLK frequency is 35 MHz for write mode and 16 MHz for readback.

Rev. F | Page 5 of 27
AD5791 Data Sheet
t1 t7

SCLK 1 2 24

t6 t3 t2
t4 t5

SYNC

t9
t8

SDIN DB23 DB0

t10 t12
t11

LDAC

VOUT t13

VOUT t14

t15

CLR

t16

VOUT

t21
RESET

t22

08964-002
VOUT

Figure 2. Write Mode Timing Diagram

t17 t1 t7 t20

SCLK 1 2 24 1 2 24

t6 t3 t2
t17 t5
t4 t5

SYNC

t9
t8

SDIN DB23 DB0

INPUT WORD SPECIFIES NOP CONDITION t18

REGISTER TO BE READ
t19

SDO DB23 DB0

08964-003

REGISTER CONTENTS CLOCKED OUT

Figure 3. Readback Mode Timing Diagram

Rev. F | Page 6 of 27
Data Sheet AD5791

ABSOLUTE MAXIMUM RATINGS

TA = 25°C, unless otherwise noted. Transient currents of up to Stresses at or above those listed under Absolute Maximum
100 mA do not cause SCR latch-up. Ratings may cause permanent damage to the product. This is a
Table 5. stress rating only; functional operation of the product at these
or any other conditions above those indicated in the operational
Parameter Rating
section of this specification is not implied. Operation beyond
VDD to AGND −0.3 V to +34 V
the maximum operating conditions for extended periods may
VSS to AGND −34 V to +0.3 V
affect product reliability.
VDD to VSS −0.3 V to +34 V
VCC to DGND −0.3 V to +7 V This device is a high performance integrated circuit with an
IOVCC to DGND −0.3 V to VCC + 0.3 V or +7 V ESD rating of 1.5 kV, and it is ESD sensitive. Take proper
(whichever is less) precautions for handling and assembly.
Digital Inputs to DGND −0.3 V to IOVCC + 0.3 V or
+7 V (whichever is less)
VOUT to AGND −0.3 V to VDD + 0.3 V ESD CAUTION
VREFPF to AGND −0.3 V to VDD + 0.3 V
VREFPS to AGND −0.3 V to VDD + 0.3 V
VREFNF to AGND VSS − 0.3 V to + 0.3 V
VREFNS to AGND VSS − 0.3 V to + 0.3 V
DGND to AGND −0.3 V to +0.3 V
Operating Temperature Range, TA
Industrial −40°C to + 125°C
Storage Temperature Range −65°C to +150°C
Maximum Junction Temperature, 150°C
TJ max
Power Dissipation (TJ max − TA)/θJA
TSSOP Package
θJA Thermal Impedance 143°C/W
θJC Thermal Impedance 45°C/W
Lead Temperature JEDEC industry standard
Soldering J-STD-020
ESD (Human Body Model) 1.5 kV

Rev. F | Page 7 of 27
AD5791 Data Sheet

PIN CONFIGURATION AND FUNCTION DESCRIPTIONS

INV 1 20 RFB

VOUT 2 19 AGND
VREFPS 3 18 VSS
AD5791
VREFPF 4 TOP VIEW 17 VREFNS
(Not to Scale) 16
VDD 5 VREFNF
RESET 6 15 DGND
CLR 7 14 SYNC

LDAC 8 13 SCLK
VCC 9 12 SDIN

08964-005
IOVCC 10 11 SDO

Figure 4. Pin Configuration

Table 6. Pin Function Descriptions

Pin No. Mnemonic Description
1 INV Connection to Inverting Input of External Amplifier. See the AD5791 Features section for further details.
2 VOUT Analog Output Voltage.
3 VREFPS Positive Reference Sense Voltage Input. A voltage range of 5 V to VDD − 2.5 V can be connected. A unity-gain
amplifier must be connected at this pin in conjunction with the VREFPF pin. See the AD5791 Features section for
further details.
4 VREFPF Positive Reference Force Voltage Input. A voltage range of 5 V to VDD − 2.5 V can be connected. A unity-gain
amplifier must be connected at this pin in conjunction with the VREFPS pin. See the AD5791 Features section for
further details.
5 VDD Positive Analog Supply Connection. A voltage range of 7.5 V to 16.5 V can be connected, decouple VDD to AGND.
6 RESET Active Low Reset Logic Input Pin. Asserting this pin returns the AD5791 to its power-on status.
7 CLR Active Low Clear Logic Input Pin. Asserting this pin sets the DAC register to a user defined value (see Table 13) and
updates the DAC output. The output value depends on the DAC register coding that is being used, either binary
or twos complement.
8 LDAC Active Low Load DAC Logic Input Pin. This is used to update the DAC register and consequently, the analog
output. When tied permanently low, the output is updated on the rising edge of SYNC. If LDAC is held high during
the write cycle, the input register is updated, but the output update is held off until the falling edge of LDAC. Do
not leave the LDAC pin unconnected.
9 VCC Digital Supply Connection. A voltage range of 2.7 V to 5.5 V can be connected. Decouple VCC to DGND.
10 IOVCC Digital Interface Supply Pin. Digital threshold levels are referenced to the voltage applied to this pin. A voltage in
the range of 1.71 V to 5.5 V can be connected. Do not allow IOVCC to exceed VCC.
11 SDO Serial Data Output Pin. Data is clocked out on the rising edge of the serial clock input.
12 SDIN Serial Data Input Pin. This device has a 24-bit shift register. Data is clocked into the register on the falling edge of
the serial clock input.
13 SCLK Serial Clock Input. Data is clocked into the input shift register on the falling edge of the serial clock input. Data can
be transferred at clock rates of up to 35 MHz.
14 SYNC Active Low Digital Interface Synchronization Input Pin. This is the frame synchronization signal for the input data.
When SYNC is low, it enables the input shift register, and data is then transferred in on the falling edges of the
following clocks. The input shift register is updated on the rising edge of SYNC.
15 DGND Ground Reference Pin for Digital Circuitry.
16 VREFNF Negative Reference Force Voltage Input. A voltage range of VSS + 2.5 V to 0 V can be connected. A unity-gain
amplifier must be connected at this pin, in conjunction with the VREFNS pin. See the AD5791 Features section for
further details.
17 VREFNS Negative Reference Sense Voltage Input. A voltage range of VSS + 2.5 V to 0 V can be connected. A unity-gain
amplifier must be connected at this pin, in conjunction with the VREFNF pin. See the AD5791 Features section for
further details.
18 VSS Negative Analog Supply Connection. A voltage range of −16.5 V to −2.5 V can be connected. Decouple VSS to AGND.
19 AGND Ground Reference Pin for Analog Circuitry.
20 RFB Feedback Connection for External Amplifier. See the AD5791 Features section for further details.

Rev. F | Page 8 of 27
Data Sheet AD5791

TYPICAL PERFORMANCE CHARACTERISTICS

1.0 0.8
AD8676 REFERENCE BUFFERS TA = +125°C AD8676 REFERENCE BUFFERS
0.8 AD8675 OUTPUT BUFFER TA = +25°C AD8675 OUTPUT BUFFER
TA = –40°C 0.6
0.6
0.4
0.4
INL ERROR (LSB)

INL ERROR (LSB)

0.2 0.2

0 0

–0.2
–0.2
–0.4
–0.4
–0.6 VREFP = +10V VREFP = +10V
VREFN = –10V VREFN = 0V TA = –40°C
–0.6
–0.8 VDD = +15V VDD = +15V TA = +125°C
VSS = –15V VSS = –15V TA = +25°C
–1.0 –0.8

08964-006

08964-009
0 200000 400000 600000 800000 1000000 0 200000 400000 600000 800000 1000000
DAC CODE DAC CODE

Figure 5. Integral Nonlinearity Error vs. DAC Code, ±10 V Span Figure 8. Integral Nonlinearity Error vs. DAC Code, ±10 V Span, X2 Gain Mode

1.5 1.0
TA = +125°C AD8676 REFERENCE BUFFERS AD8676 REFERENCE BUFFERS VREFP = +10V
TA = +25°C AD8675 OUTPUT BUFFER 0.8 AD8675 OUTPUT BUFFER VREFN = –10V
TA = –40°C
VDD = +15V
1.0
0.6 VSS = –15V

0.4
0.5 DNL ERROR (LSB)
INL ERROR (LSB)

0.2

0 0

–0.2
–0.5
–0.4

VREFP = +10V –0.6

–1.0 VREFN = 0V TA = +125°C
VDD = +15V –0.8 TA = +25°C
VSS = –15V TA = –40°C
–1.5 –1.0
08964-007

08964-010
0 200000 400000 600000 800000 1000000 0 200000 400000 600000 800000 1000000
DAC CODE DAC CODE
Figure 6. Integral Nonlinearity Error vs. DAC Code, 10 V Span Figure 9. Differential Nonlinearity Error vs. DAC Code, ±10 V Span

2.5 1.5
TA = +125°C AD8676 REFERENCE BUFFERS VREFP = +10V
AD8675 OUTPUT BUFFER VREFN = 0V
2.0 TA = +25°C VDD = +15V
TA = –40°C VSS = –15V
1.0
1.5

1.0
0.5
DNL ERROR (LSB)
INL ERROR (LSB)

0.5

0 0

–0.5
–0.5
–1.0

–1.5 VREFP = +5V

–1.0
VREFN = 0V TA = +125°C
–2.0 VDD = +15V AD8676 REFERENCE BUFFERS TA = +25°C
VSS = –15V AD8675 OUTPUT BUFFER TA = –40°C
–2.5 –1.5
08964-011
08964-008

0 200000 400000 600000 800000 1000000 0 200000 400000 600000 800000 1000000
DAC CODE DAC CODE

Figure 7. Integral Nonlinearity Error vs. DAC Code, 5 V Span Figure 10. Differential Nonlinearity Error vs. DAC Code, 10 V Span

Rev. F | Page 9 of 27
AD5791 Data Sheet
2.0 1.0
TA = +125°C VREFP = +5V ±10V SPAN MAX DNL +10V SPAN MAX DNL
TA = +25°C VREFN = 0V +5V SPAN MAX DNL ±10V SPAN MIN DNL
1.5 TA = –40°C VDD = +15V +10V SPAN MIN DNL +5V SPAN MIN DNL
VSS = –15V 0.5
1.0

DNL ERROR (LSB)

DNL ERROR (LSB)

0.5
0

–0.5
–0.5

–1.0 AD8676 REFERENCE BUFFERS

–1.0 AD8675 OUTPUT BUFFER
–1.5 VDD = +15V
AD8676 REFERENCE BUFFERS VSS = –15V
AD8675 OUTPUT BUFFER
–2.0 –1.5

08964-015
08964-012
0 200000 400000 600000 800000 1000000 –55 –35 –15 5 25 45 65 85 105 125
DAC CODE TEMPERATURE (°C)

Figure 11. Differential Nonlinearity Error vs. DAC Code, 5 V Span Figure 14. Differential Nonlinearity Error vs. Temperature

1.0 0.6
AD8676 REFERENCE BUFFERS TA = +25°C
0.8 AD8675 OUTPUT BUFFER TA = –40°C
0.5
INL MAX
VREFP = +10V TA = +125°C
0.6 VREFN = 0V
VDD = +15V 0.4
0.4 VSS = –15V
DNL ERROR (LSB)

INL ERROR (LSB) 0.3

0.2 TA = 25°C
0.2 VREFP = +10V
0 VREFN = –10V
AD8676 REFERENCE BUFFERS
0.1
–0.2 AD8675 OUTPUT BUFFER

0
–0.4
INL MIN
–0.6 –0.1

–0.8 –0.2

–1.0 –0.3
08964-013

08964-016
0 200000 400000 600000 800000 1000000 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5
DAC CODE VDD/|VSS| (V)

Figure 12. Differential Nonlinearity Error vs. DAC Code, ±10 V Span, Figure 15. Integral Nonlinearity Error vs. Supply Voltage, ±10 V Span
X2 Gain Mode
2.0 1.5
±10V SPAN MAX INL +10V SPAN MAX INL
+5V SPAN MAX INL ±10V SPAN MIN INL
1.5 INL MAX
+10V SPAN MIN INL +5V SPAN MIN INL 1.0

1.0
TA = 25°C
INL ERROR (LSB)

0.5
INL ERROR (LSB)

VREFP = +5V
0.5 VREFN = 0V
AD8676 REFERENCE BUFFERS
0 AD8675 OUTPUT BUFFER
0

–0.5
–0.5

AD8676 REFERENCE BUFFERS –1.0

–1.0 AD8675 OUTPUT BUFFER INL MIN
VDD = +15V
VSS = –15V
–1.5 –1.5
08964-014

–55 –35 –15 5 25 45 65 85 105 125 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5
TEMPERATURE (°C) VDD (V)
08964-017

–2.5 –3.9 –5.3 –6.7 –9.1 –10.5 –12.9 –14.2 –15.5 –16.5
VSS (V)

Figure 13. Integral Nonlinearity Error vs. Temperature Figure 16. Integral Nonlinearity Error vs. Supply Voltage, 5 V Span

Rev. F | Page 10 of 27
Data Sheet AD5791
0.4 0.6
TA = 25°C
DNL MAX VREFP = +5V
0.3
0.5 VREFN = 0V
AD8676 REFERENCE BUFFERS

ZERO-SCALE ERROR (LSB)

0.2 AD8675 OUTPUT BUFFER
DNL ERROR (LSB)

0.4
0.1 TA = 25°C
VREFP = +10V
0 VREFN = –10V 0.3
AD8676 REFERENCE BUFFERS
AD8675 OUTPUT BUFFER
–0.1
0.2

–0.2
0.1
–0.3
DNL MIN

–0.4 0

08964-018
12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5
VDD/|VSS| (V) VDD (V)

08964-021
–2.5 –3.9 –5.3 –6.7 –9.1 –10.5 –12.9 –14.2 –15.5 –16.5
VSS (V)

Figure 17. Differential Nonlinearity Error vs. Supply Voltage, ±10 V Span Figure 20. Zero-Scale Error vs. Supply Voltage, 5 V Span

0.4 0.20
TA = 25°C
DNL MAX VREFP = +10V
0.2 0.15 VREFN = –10V
AD8676 REFERENCE BUFFERS
AD8675 OUTPUT BUFFER

MIDSCALE ERROR (LSB)

0 0.10
DNL ERROR (LSB)

TA = 25°C
–0.2 VREFP = +5V 0.05
VREFN = 0V
AD8676 REFERENCE BUFFERS
–0.4 AD8675 OUTPUT BUFFER 0

–0.6 –0.05
DNL MIN

–0.8 –0.10

–1.0 –0.15

08964-022
7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5 12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5
VDD (V) VDD/|VSS| (V)
08964-019

–2.5 –3.9 –5.3 –6.7 –9.1 –10.5 –12.9 –14.2 –15.5 –16.5
VSS (V)

Figure 18. Differential Nonlinearity Error vs. Supply Voltage, 5 V Span Figure 21. Midscale Error vs. Supply Voltage, ±10 V Span

0.6 0.2

0.1
0.5
0
ZERO-SCALE ERROR (LSB)

MIDSCALE ERROR (LSB)

0.4 –0.1

–0.2
0.3
TA = 25°C –0.3
VREFP = +10V
0.2 VREFN = –10V –0.4
AD8676 REFERENCE BUFFERS TA = 25°C
AD8675 OUTPUT BUFFER –0.5 VREFP = +5V
0.1 VREFN = 0V
–0.6 AD8676 REFERENCE BUFFERS
AD8675 OUTPUT BUFFER
0 –0.7
08964-020

12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5
VDD/|VSS| (V) VDD (V)
08964-023

–2.5 –3.9 –5.3 –6.7 –9.1 –10.5 –12.9 –14.2 –15.5 –16.5
VSS (V)

Figure 19. Zero-Scale Error vs. Supply Voltage, ±10 V Span Figure 22. Midscale Error vs. Supply Voltage, 5 V Span

Rev. F | Page 11 of 27
AD5791 Data Sheet
–0.015 0.10
TA = 25°C
TA = 25°C
–0.035 VREFP = +10V 0.05 VREFP = +5V
VREFN = –10V
VREFN = 0V
–0.055 AD8676 REFERENCE BUFFERS 0 AD8676 REFERENCE BUFFERS
FULL-SCALE ERROR (LSB)

AD8675 OUTPUT BUFFER

AD8675 OUTPUT BUFFER

GAIN ERROR (ppm FSR)

–0.075 –0.05

–0.10
–0.095
–0.15
–0.115
–0.20
–0.135
–0.25
–0.155
–0.30

–0.175 –0.35

–0.195 –0.40

08964-024
12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5
VDD/|VSS| (V) VDD (V)

08964-027
–2.5 –3.9 –5.3 –6.7 –9.1 –10.5 –12.9 –14.2 –15.5 –16.5
VSS (V)

Figure 23. Full-Scale Error vs. Supply Voltage, ±10 V Span Figure 26. Gain Error vs. Supply Voltage, 5 V Span

0.25 0.6

0.20 0.4 INL MAX

FULL-SCALE ERROR (LSB)

0.15 0.2
INL ERROR (LSB)

TA = 25°C
VDD = +15V
VSS = –15V
0.10 0
AD8676 REFERENCE BUFFERS
AD8675 OUTPUT BUFFER

0.05 TA = 25°C –0.2

VREFP = +5V
VREFN = 0V
0 AD8676 REFERENCE BUFFERS
–0.4
AD8675 OUTPUT BUFFER INL MIN

–0.05 –0.6

08964-028
7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0
VDD (V) VREFP /|VREFN | (V)
08964-025

–2.5 –3.9 –5.3 –6.7 –9.1 –10.5 –12.9 –14.2 –15.5 –16.5
VSS (V)

Figure 24. Full-Scale Error vs. Supply Voltage, 5 V Span Figure 27. Integral Nonlinearity Error vs. Reference Voltage

–0.30 0.4
TA = 25°C
DNL MAX
VREFP = +10V 0.3
–0.35 VREFN = –10V
AD8676 REFERENCE BUFFERS 0.2
AD8675 OUTPUT BUFFER
GAIN ERROR (ppm FSR)

–0.40 0.1
DNL ERROR (LSB)

0 TA = 25°C
–0.45 VDD = +15V
–0.1 VSS = –15V
AD8676 REFERENCE BUFFERS
–0.50
–0.2 AD8675 OUTPUT BUFFER

–0.55 –0.3

–0.4
–0.60
–0.5
DNL MIN
–0.65 –0.6
08964-026

08964-029

12.5 13.0 13.5 14.0 14.5 15.0 15.5 16.0 16.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0
VDD/|VSS| (V) VREFP /|VREFN | (V)

Figure 25. Gain Error vs. Supply Voltage, ±10 V Span Figure 28. Differential Nonlinearity Error vs. Reference Voltage

Rev. F | Page 12 of 27
Data Sheet AD5791
0.60 –0.30
TA = 25°C
VDD = +15V
0.55 –0.35 VSS = –15V
AD8676 REFERENCE BUFFERS
ZERO-SCALE ERROR (LSB)

AD8675 OUTPUT BUFFER

GAIN ERROR (ppm FSR)

0.50 –0.40

0.45 –0.45
TA = 25°C
VDD = +15V
0.40 VSS = –15V –0.50
AD8676 REFERENCE BUFFERS
AD8675 OUTPUT BUFFER
0.35 –0.55

0.30 –0.60

08964-030

08964-033
5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0
VREFP /|VREFN | (V) VREFP /|VREFN | (V)

Figure 29. Zero-Scale Error vs. Reference Voltage Figure 32. Gain Error vs. Reference Voltage

0.15 2.0
AD8676 REFERENCE BUFFERS ±10V SPAN
1.5 AD8675 OUTPUT BUFFER +10V SPAN
0.10 VDD = +15V +5V SPAN
1.0 VSS = –15V

FULL-SCALE ERROR (LSBs)

MIDSCALE ERROR (LSB)

0.05 0.5

0
0
–0.5
–0.05 TA = 25°C
VDD = +15V –1.0
VSS = –15V
–0.10 –1.5
AD8676 REFERENCE BUFFERS
AD8675 OUTPUT BUFFER
–2.0
–0.15
–2.5

–0.20 –3.0

08964-034
08964-031

5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 –55 –35 –15 5 25 45 65 85 105 125
VREFP /|VREFN | (V) TEMPERATURE (°C)

Figure 30. Midscale Error vs. Reference Voltage Figure 33. Full-Scale Error vs. Temperature

0.15 2.0
±10V SPAN
1.8 +10V SPAN
0.10 +5V SPAN
1.6
FULL-SCALE ERROR (LSB)

MIDSCALE ERROR (LSBs)

0.05 1.4

1.2
0
1
–0.05
0.8
TA = 25°C
VDD = +15V 0.6
–0.10
VSS = –15V AD8676 REFERENCE BUFFERS
0.4
AD8676 REFERENCE BUFFERS AD8675 OUTPUT BUFFER
–0.15 AD8675 OUTPUT BUFFER
0.2 VDD = +15V
VSS = –15V
–0.20 0
08964-035
08964-032

5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 –55 –35 –15 5 25 45 65 85 105 125
VREFP /|VREFN | (V) TEMPERATURE (°C)

Figure 31. Full-Scale Error vs. Reference Voltage Figure 34. Midscale Error vs. Temperature

Rev. F | Page 13 of 27
AD5791 Data Sheet
5 5
±10V SPAN TA = 25°C
4 +10V SPAN 4
+5V SPAN IDD
3 3
ZERO-SCALE ERROR (LSBs)

2 2

IDD, ISS (mA)

1 1

0 0

–1 –1

–2 –2

–3 AD8676 REFERENCE BUFFERS –3

AD8675 OUTPUT BUFFER ISS
–4 VDD = +15V
–4
VSS = –15V
–5 –5

08964-039
08964-036
–55 –35 –15 5 25 45 65 85 105 125 –20 –15 –10 –5 0 5 10 15 20
TEMPERATURE (°C) VDD/VSS (V)

Figure 35. Zero-Scale Error vs. Temperature Figure 38. Power Supply Currents vs. Power Supply Voltages

4
AD8676 REFERENCE BUFFERS ±10V SPAN
3 AD8675 OUTPUT BUFFER +10V SPAN
VDD = +15V +5V SPAN
VSS = –15V
2
GAIN ERROR (ppm FSR)

1
VDD = +15V
VSS = –15V
0 VREFP = +10V
3 VREFN = –10V
–1 AD8676 REFERENCE BUFFERS
OUTPUT UNBUFFERED
–2 LOAD = 10MΩ||20pF

–3

–4

08964-040
–5 4
08964-037

–55 –35 –15 5 25 45 65 85 105 125 CH3 5V CH4 5V 200ns

TEMPERATURE (°C)

Figure 36. Gain Error vs. Temperature Figure 39. Rising Full-Scale Voltage Step

900
TA = 25°C IOVCC = 5V, LOGIC VOLTAGE
VDD = +15V
INCREASING
800 VSS = –15V
IOVCC = 5V, LOGIC VOLTAGE
VREFP = +10V
DECREASING
VREFN = –10V
700 IOVCC = 3V, LOGIC VOLTAGE
INCREASING AD8676 REFERENCE BUFFERS
OUTPUT UNBUFFERED
600 IOVCC = 3V, LOGIC VOLTAGE
LOAD = 10MΩ||20pF
DECREASING
IOICC (µA)

500
3
400

300

200

100
08964-041

4
0
08964-038

CH3 5V CH4 5V 200ns

0 1 2 3 4 5 6
LOGIC INPUT VOLTAGE (V)

Figure 37. IOICC vs. Logic Input Voltage Figure 40. Falling Full-Scale Voltage Step

Rev. F | Page 14 of 27
Data Sheet AD5791
10.8 3.0
±10V VREF 5V VREF
OUTPUT GAIN OF 1 OUTPUT GAIN OF 1
2.6
10.6 BIAS COMPENSATION MODE BIAS COMPENSATION MODE
20pF COMPENSATION CAPACITOR 20pF COMPENSATION CAPACITOR

OUTPUT GLITCH (nV–sec)

RC LOW-PASS FILTER 2.2 RC LOW-PASS FILTER NEGATIVE CODE
10.4 CHANGE
1.8 POSITIVE CODE
CHANGE
VOUT (mV)

10.2
1.4
10.0
1.0

9.8
0.6

9.6 0.2

088964-063
9.4 –0.2

114688
163840
212992
262144
311296
360448
409600
458752
507904
557056
606208
655360
704512
753664
802816
851968
901120
950272
999424
0 1 2 3 4 5

65536
16384
TIME (µs)

08964-061
CODE

Figure 41. 500 Code Step Settling Time Figure 44. 6 MSB Segment Glitch Energy for +5 V VREF

10 40
5V VREF ±10V VREF
9 OUTPUT GAIN OF 1 NEGATIVE CODE OUTPUT GAIN OF 1
BIAS COMPENSATION MODE CHANGE BIAS COMPENSATION MODE
30
8 20pF COMPENSATION CAPACITOR 20pF COMPENSATION CAPACITOR
OUTPUT GLITCH (nV–sec)

RC LOW-PASS FILTER RC LOW-PASS FILTER

7
20
6
VOUT (mV)

5 POSITIVE CODE 10
CHANGE
4
0
3

2 CX = 143pF + 0pF
–10 CX = 143pF + 220pF
1 CX = 143pF + 470pF
CX = 143pF + 1,000pF

088964-062
0 –20
114688
163840
212992
262144
311296
360448
409600
458752
507904
557056
606208
655360
704512
753664
802816
851968
901120
950272
999424
65536
16384

–1.0 –0.5 0 0.5 1.0 1.5 2.0

TIME (µs)
08964-059

CODE

Figure 42. 6 MSB Segment Glitch Energy for ±10 V VREF Figure 45. Midscale Peak-to-Peak Glitch for ±10 V

4.0 800
10V VREF TA = 25°C MIDSCALE CODE LOADED
OUTPUT GAIN OF 1 VDD = +15V OUTPUT UNBUFFERED
3.5 V
BIAS COMPENSATION MODE 600 SS = –15V AD8676 REFERENCE BUFFERS
20pF COMPENSATION CAPACITOR VREFP = +10V
POSITIVE CODE
OUTPUT GLITCH (nV–sec)

3.0 RC LOW-PASS FILTER VREFN = –10V

CHANGE
OUTPUT VOLTAGE (nV)

400
NEGATIVE CODE
CHANGE
2.5
200
2.0
0
1.5

–200
1.0

0.5 –400

0 –600
08964-044

0 1 2 3 4 5 6 7 8 9 10
114688
163840
212992
262144
311296
360448
409600
458752
507904
557056
606208
655360
704512
753664
802816
851968
901120
950272
999424
65536
16384

TIME (Seconds)
08964-060

CODE

Figure 43. 6 MSB Segment Glitch Energy for +10 V VREF Figure 46. Voltage Output Noise, 0.1 Hz to 10 Hz Bandwidth

Rev. F | Page 15 of 27
AD5791 Data Sheet
100 350
VDD = +15V
TA = 25°C
VSS = –15V
300 VDD = +15V
VREFP = +10V
VSS = –15V
VREFN = –10V
VREFP = +10V
CODE = MIDSCALE 250 VREFN = –10V

OUTPUT VOLTAGE (mV)

AD8675 OUTPUT BUFFER
200
NSD (nV/ Hz)

10 150

100

50

1 –50

08964-049
08964-064
0.1 1 10 100 1k 10k 100k –1 0 1 2 3 4 5 6
FREQUENCY (Hz) TIME (µs)

Figure 47. Noise Spectral Density vs. Frequency Figure 48. Glitch Impulse on Removal of Output Clamp

Rev. F | Page 16 of 27
Data Sheet AD5791

TERMINOLOGY
Relative Accuracy Midscale Error Temperature Coefficient
Relative accuracy, or integral nonlinearity (INL), is a measure of Midscale error temperature coefficient is a measure of the
the maximum deviation, in LSB, from a straight line passing change in midscale error with a change in temperature. It is
through the endpoints of the DAC transfer function. A typical expressed in ppm FSR/°C.
INL error vs. code plot is shown in Figure 5. Output Slew Rate
Differential Nonlinearity (DNL) Slew rate is a measure of the limitation in the rate of change of
Differential nonlinearity is the difference between the measured the output voltage. The slew rate of the AD5791 output voltage
change and the ideal 1 LSB change between any two adjacent is determined by the capacitive load presented to the VOUT pin. The
codes. A specified differential nonlinearity of ±1 LSB maximum capacitive load in conjunction with the 3.4 kΩ output impedance
ensures monotonicity. This DAC is guaranteed monotonic. A of the AD5791 set the slew rate. Slew rate is measured from 10%
typical DNL error vs. code plot is shown in Figure 9. to 90% of the output voltage change and is expressed in V/µs.
Linearity Error Long-Term Stability Output Voltage Settling Time
Linearity error long-term stability is a measure of the stability of Output voltage settling time is the amount of time it takes for
the linearity of the DAC over a long period. It is specified in LSB the output voltage to settle to a specified level for a specified
for a time period of 500 hours and 1000 hours at an elevated change in voltage. For fast settling applications, a high speed
ambient temperature. buffer amplifier is required to buffer the load from the 3.4 kΩ
Zero-Scale Error output impedance of the AD5791, in which case it is the
Zero-scale error is a measure of the output error when zero-scale amplifier that determines the settling time.
code (0x00000) is loaded to the DAC register. Ideally, the output Digital-to-Analog Glitch Impulse
voltage is VREFNS. Zero-scale error is expressed in LSBs. Digital-to-analog glitch impulse is the impulse injected into the
Zero-Scale Error Temperature Coefficient analog output when the input code in the DAC register changes
Zero-scale error temperature coefficient is a measure of the state. It is specified as the area of the glitch in nV-sec and is
change in zero-scale error with a change in temperature. It is measured when the digital input code is changed by 1 LSB at
expressed in ppm FSR/°C. the major carry transition (see Figure 42).

Full-Scale Error Output Enabled Glitch Impulse

Full-scale error is a measure of the output error when full- Output enabled glitch impulse is the impulse injected into the
scale code (0x3FFFF) is loaded to the DAC register. Ideally, analog output when the clamp to ground on the DAC output is
the output voltage is VREFPS − 1 LSB. Full-scale error is expressed in removed. It is specified as the area of the glitch in nV-sec (see
LSBs. Figure 48).

Full-Scale Error Temperature Coefficient Digital Feedthrough

Full-scale error temperature coefficient is a measure of the Digital feedthrough is a measure of the impulse injected into
change in full-scale error with a change in temperature. It is the analog output of the DAC from the digital inputs of the
expressed in ppm FSR/°C. DAC but is measured when the DAC output is not updated. It is
specified in nV-sec and measured with a full-scale code change
Gain Error on the data bus, that is, from all 0s to all 1s, and vice versa.
Gain error is a measure of the span error of the DAC. It is the
deviation in slope of the DAC transfer characteristic from ideal, Spurious Free Dynamic Range (SFDR)
expressed in ppm of the full-scale range. Spurious free dynamic range is the usable dynamic range of a
Gain Error Temperature Coefficient DAC before spurious noise interferes or distorts the fundamental
Gain error temperature coefficient is a measure of the change in signal. It is measured by the difference in amplitude between
gain error with a change in temperature. It is expressed in ppm the fundamental and the largest harmonically or nonharmonically
FSR/°C. related spur from dc to full Nyquist bandwidth (half the DAC
sampling rate, or fS/2). SFDR is measured when the signal is a
Midscale Error digitally generated sine wave.
Midscale error is a measure of the output error when midscale
code (0x20000) is loaded to the DAC register. Ideally, the output Total Harmonic Distortion (THD)
voltage is (VREFPS − VREFNS)/2 +VREFNS. Midscale error is Total harmonic distortion is the ratio of the rms sum of the
expressed in LSBs. harmonics of the DAC output to the fundamental value. Only
the second to fifth harmonics are included.

Rev. F | Page 17 of 27
AD5791 Data Sheet
DC Power Supply Rejection Ratio AC Power Supply Rejection Ratio (AC PSRR)
DC power supply rejection ratio is a measure of the rejection of AC power supply rejection ratio is a measure of the rejection of
the output voltage to dc changes in the power supplies applied the output voltage to ac changes in the power supplies applied
to the DAC. This ratio is measured for a given dc change in to the DAC. This ratio is measured for a given amplitude and
power supply voltage and is expressed in µV/V. frequency change in power supply voltage and is expressed in
decibels.

Rev. F | Page 18 of 27
Data Sheet AD5791

THEORY OF OPERATION
R R R
The AD5791 is a high accuracy, fast settling, single, 20-bit, VOUT

serial input, voltage output DAC that operates from a VDD supply 2R 2R 2R ..................... 2R 2R 2R .......... 2R

voltage of 7.5 V to 16.5 V and a VSS supply of −16.5 V to −2.5 V. S0 S1 ..................... S13 E62 E61.......... E0
VREFPF
Data is written to the AD5791 in a 24-bit word format via a 3-wire VREFPS
serial interface. The AD5791 incorporates a power-on reset circuit VREFNF
VREFNS
that ensures the DAC output powers up to 0 V with the VOUT pin

08964-050
14-BIT R-2R LADDER SIX MSBs DECODED INTO
clamped to AGND through a ~6 kΩ internal resistor. 63 EQUAL SEGMENTS

DAC ARCHITECTURE Figure 49. DAC Ladder Structure

The architecture of the AD5791 consists of two matched DAC SERIAL INTERFACE
sections. Figure 49 shows a simplified circuit diagram. The The AD5791 has a 3-wire serial interface (SYNC, SCLK, and
6 MSBs of the 20-bit data-word are decoded to drive 63 switches,
SDIN) that is compatible with SPI, QSPI, and MICROWIRE
E0 to E62. Each of these switches connects one of 63 matched
interface standards, as well as most DSPs (see Figure 2 for a
resistors to either the VREFP or VREFN voltage. The remaining
timing diagram).
14 bits of the data-word drive the S0 to S13 switched of a 14-bit
voltage mode R-2R ladder network. To ensure performance to Input Shift Register
specification, the reference inputs must be force sensed with The input shift register is 24 bits wide. Data is loaded into the
external amplifiers. device MSB first as a 24-bit word under the control of a serial
clock input, SCLK, which can operate at up to 35 MHz. The
input register consists of an R/W bit, three address bits, and
twenty register bits as shown in Table 7. The timing diagram for
this operation is shown in Figure 2.

Table 7. Input Shift Register Format

MSB LSB
DB23 DB22 DB21 DB20 DB19 DB0
R/W Register address Register data

Table 8. Decoding the Input Shift Register

R/W Register Address Description
X1 0 0 0 No operation (NOP; used in readback operations
0 0 0 1 Write to the DAC register
0 0 1 0 Write to the control register
0 0 1 1 Write to the clearcode register
0 1 0 0 Write to the software control register
1 0 0 1 Read from the DAC register
1 0 1 0 Read from the control register
1 0 1 1 Read from the clearcode register
1
X is don’t care.

Rev. F | Page 19 of 27
AD5791 Data Sheet
Standalone Operation HARDWARE CONTROL PINS
The serial interface works with both a continuous and noncon- Load DAC Function (LDAC)
tinuous serial clock. A continuous SCLK source can be used After data has been transferred into the input register of the
only if SYNC is held low for the correct number of clock cycles. DAC, there are two ways to update the DAC register and DAC
In gated clock mode, a burst clock containing the exact number output. Depending on the status of both SYNC and LDAC, one
of clock cycles must be used, and SYNC must be taken high of two update modes is selected: synchronous DAC updating or
after the final clock to latch the data. The first falling edge of asynchronous DAC updating
SYNC starts the write cycle. Exactly 24 falling clock edges must
Synchronous DAC Update
be applied to SCLK before SYNC is brought high again. If
SYNC is brought high before the 24th falling SCLK edge, the In this mode, LDAC is held low while data is being clocked into
data written is invalid. If more than 24 falling SCLK edges are the input shift register. The DAC output is updated on the rising
applied before SYNC is brought high, the input data is also edge of SYNC.
invalid. The input shift register is updated on the rising edge of Asynchronous DAC Update
SYNC. For another serial transfer to take place, SYNC must be
In this mode, LDAC is held high while data is being clocked
brought low again. After the end of the serial data transfer, data
into the input shift register. The DAC output is asynchronously
is automatically transferred from the input shift register to the
updated by taking LDAC low after SYNC has been taken high.
addressed register. When the write cycle is complete, the output
can be updated by taking LDAC low while SYNC is high. The update now occurs on the falling edge of LDAC.

Readback Reset Function (RESET)

The contents of all the on-chip registers can be read back via the The AD5791 can be reset to its power-on state by two means:
SDO pin. Table 8 outlines how the registers are decoded. After a either by asserting the RESET pin or by utilizing the software
register has been addressed for a read, the next 24 clock cycles RESET control function (see Table 14). If the RESET pin is not
clock the data out on the SDO pin. The clocks must be applied used, hardwire it to IOVCC.
while SYNC is low. When SYNC is returned high, the SDO pin Asynchronous Clear Function (CLR)
is placed in tristate. For a read of a single register, the NOP
The CLR pin is an active low clear that allows the output to
function can be used to clock out the data. Alternatively, if more
be cleared to a user defined value. The 20-bit clear code value
than one register is to be read, the data of the first register to be
is programmed to the clearcode register (see Table 13). It is
addressed can be clocked out at the same time the second register
necessary to maintain CLR low for a minimum amount of time
to be read is being addressed. The SDO pin must be enabled to
complete a readback operation. The SDO pin is enabled by to complete the operation (see Figure 2).When the CLR signal
default. is returned high the output remains at the clear value (if LDAC
is high) until a new value is loaded to the DAC register. The
output cannot be updated with a new value while the CLR pin is
low. A clear operation can also be performed by setting the CLR
bit in the software control register (see Table 14).

Rev. F | Page 20 of 27
Data Sheet AD5791
Table 9. Hardware Control Pins Truth Table
LDAC CLR RESET Function
X1 X1 0 The AD5791 is in reset mode. The device cannot be programmed.
X1 X1 The AD5791 is returned to its power-on state. All registers are set to their default values.
0 0 1 The DAC register is loaded with the clearcode register value and the output is set accordingly.
0 1 1 The output is set according to the DAC register value.
1 0 1 The DAC register is loaded with the clearcode register value and the output is set accordingly.
1 1 The output is set according to the DAC register value.
0 1 The output remains at the clear code value.
1 1 The output remains set according to the DAC register value.
0 1 The output remains at the clear code value.
1 1 The DAC register is loaded with the clearcode register value and the output is set accordingly.
0 1 The DAC register is loaded with the clearcode register value and the output is set accordingly.
1 1 The output remains at the clear code value
0 1 The output is set according to the DAC register value.
1
X is don’t care.

ON-CHIP REGISTERS
DAC Register
Table 10 outlines how data is written to and read from the DAC register.

Table 10. DAC Register

MSB LSB
DB23 DB22 DB21 DB20 DB19 DB0
R/W Register address DAC register data
R/W 0 0 1 20-bits of data
The following equation describes the ideal transfer function of the DAC:
(VREFP − VREFN )× D
VOUT = + VREFN
2 20 − 1
where:
VREFN is the negative voltage applied at the VREFN input pins.
VREFP is the positive voltage applied at the VREFP input pins.
D is the 20-bit code programmed to the DAC.

Rev. F | Page 21 of 27
AD5791 Data Sheet
Control Register
The control register controls the mode of operation of the AD5791.

Table 11. Control Register

MSB LSB
DB23 DB22 DB21 DB20 DB19…DB11 DB10 DB9 DB8 DB7 DB6 DB5 DB4 DB3 DB2 DB1 DB0
R/W Register address Control register data
R/W 0 1 0 Reserved Reserved LIN COMP SDODIS BIN/2sC DACTRI OPGND RBUF Reserved

Table 12. Control Register Functions

Function Description
Reserved These bits are reserved and should be programmed to zero.
RBUF Output amplifier configuration control.
0: internal amplifier, A1, is powered up and Resistor RFB and R1 are connected in series as shown in Figure 53. This allows
an external amplifier to be connected in a gain of two configurations. See the AD5791 Features section for further details.
1: (default) internal amplifier, A1, is powered down and Resistor RFB and R1 are connected in parallel as shown in Figure 52 so
that the resistance between the RFB and INV pins is 3.4 kΩ, equal to the resistance of the DAC. This allows the RFB and INV pins to
be used for input bias current compensation for an external unity-gain amplifier. See the AD5791 Features section for
further details.
OPGND Output ground clamp control.
0: DAC output clamp to ground is removed and the DAC is placed in normal mode.
1: (default) DAC output is clamped to ground through a ~6 kΩ resistance, and the DAC is placed in tristate mode.
Resetting the device puts the DAC in OPGND mode, where the output ground clamp is enabled and the DAC is tristated.
Setting the OPGND bit to 1 in the control register overrules any write to the DACTRI bit.
DACTRI DAC tristate control.
0: DAC is in normal operating mode.
1: (default) DAC is in tristate mode.
BIN/2sC DAC register coding select.
0: (default) DAC register uses twos complement coding.
1: DAC register uses offset binary coding.
SDODIS SDO pin enable/disable control.
0: (default) SDO pin is enabled.
1: SDO pin is disabled (tristate).
LIN COMP Linearity error compensation for varying reference input spans. See the AD5791 Features section for further details.
0 0 0 0 (Default) reference input span up to 10 V.
1 0 0 1 Reference input span between 10 V and 12 V.
1 0 1 0 Reference input span between 12 V and 16 V.
1 0 1 1 Reference input span between16 V and 19 V.
1 1 0 0 Reference input span between 19 V and 20 V.
R/W Read/write select bit.
0: AD5791 is addressed for a write operation.
1: AD5791 is addressed for a read operation.

Clearcode Register
The clearcode register sets the value to which the DAC output is set when the CLR pin or CLR bit is asserted. The output value depends
on the DAC coding that is being used, either binary or twos complement. The default register value is 0.

Table 13. Clearcode Register

MSB LSB
DB23 DB22 DB21 DB20 DB19 DB0
R/W Register address Clearcode register data
R/W 0 1 1 20-bits of data

Rev. F | Page 22 of 27
Data Sheet AD5791
Software Control Register
This is a write only register in which writing a 1 to a particular bit has the same effect as pulsing the corresponding pin low.

Table 14. Software Control Register

MSB LSB
DB23 DB22 DB21 DB20 DB19 DB3 DB2 DB1 DB0
R/W Register address Software control register data
0 1 0 0 Reserved RESET CLR 1 LDAC 2
1
The CLR function has no effect if the LDAC pin is low.
2
The LDAC function has no effect if the CLR pin is low.
Table 15. Software Control Register Functions
Function Description
LDAC Setting this bit to a 1 updates the DAC register and consequently the DAC output.
CLR Setting this bit to a 1 sets the DAC register to a user defined value (see Table 13) and updates the DAC output. The output
value depends on the DAC register coding that is being used, either binary or twos complement.
RESET Setting this bit to a 1 returns the AD5791 to its power-on state.

Rev. F | Page 23 of 27
AD5791 Data Sheet

AD5791 FEATURES
POWER-ON TO 0 V LINEARITY COMPENSATION
The AD5791 contains a power-on reset circuit that, as well as The integral nonlinearity (INL) of the AD5791 can vary according
resetting all registers to their default values, controls the output to the applied reference voltage span, the LIN COMP bits of
voltage during power-up. Upon power-on the DAC is placed in the control register can be programmed to compensate for
tristate (its reference inputs are disconnected) and its output is this variation in INL. The specifications in this data sheet are
clamped to ground through a ~6 kΩ resistor. The DAC remains obtained with LIN COMP = 0000 for reference spans up to
in this state until programmed otherwise via the control and including 10 V and with LIN COMP = 1100 for a reference
register. This is a useful feature in applications where it is span of 20 V. The default value of the LIN COMP bits is 0000.
important to know the state of the DAC output while it is in the Intermediate LIN COMP values can be programmed for reference
process of powering up. spans between 10 V and 20 V as shown in Table 12.
POWER-UP SEQUENCE OUTPUT AMPLIFIER CONFIGURATION
To power up the device in a known safe state, power up the VDD There are a number of different ways that an output amplifier
supply before powering up the VCC supply. This step ensures can be connected to the AD5791, depending on the voltage
that VCC does not come up while VDD is unpowered during references applied and the desired output voltage span.
power-on. If the device cannot be powered-up in a safe state, Unity-Gain Configuration
connect an external Schottky diode across the VDD and VCC
Figure 51 shows an output amplifier configured for unity gain,
supplies as shown in Figure 50.
in this configuration the output spans from VREFN to VREFP.
VCC VDD
VREFP

1/2 AD8676
VCC VDD
VREFPF VREFPS
AD5791 R1 RFB RFB
08964-150

A1
6.8kΩ 6.8kΩ
AD8675,
INV ADA4898-1
Figure 50. Schottky Diode Connection 20-BIT
DAC VOUT VOUT
CONFIGURING THE AD5791
After power-on the AD5791 must be configured to put it into AD5791
VREFNF VREFNS
normal operating mode before programming the output. To do
this, the control register must be programmed. The DAC is 1/2 AD8676
removed from tristate by clearing the DACTRI bit, and the

08964-051
output clamp is removed by clearing the OPGND bit. At this VREFN
point, the output goes to VREFN, unless an alternative value is
Figure 51. Output Amplifier in Unity-Gain Configuration
first programmed to the DAC register.
A second unity-gain configuration for the output amplifier is
DAC OUTPUT STATE
one that removes an offset from the input bias currents of the
The DAC output can be placed in one of three states, controlled amplifier. It does this by inserting a resistance in the feedback
by the DACTRI and OPGND bits of the control register, as path of the amplifier that is equal to the output resistance of the
shown in Table 16. DAC. The DAC output resistance is 3.4 kΩ, by connecting R1
and RFB in parallel, a resistance equal to the DAC resistance is
Table 16. AD5791 Output State Truth Table
available on chip. Because the resistors are all on one piece of
DACTRI OPGND Output State
silicon, they are temperature coefficient matched. To enable this
0 0 Normal operating mode
mode of operation the RBUF bit of the control register must be
0 1 Output is clamped via ~6 kΩ to AGND
set to Logic 1. Figure 52 shows how the output amplifier is
1 0 Output is in tristate
connected to the AD5791. In this configuration, the output
1 1 Output is clamped via ~6 kΩ to AGND
amplifier is in unity gain and the output spans from VREFN to
VREFP. This unity-gain configuration allows a capacitor to be placed
in the amplifier feedback path to improve dynamic performance.

Rev. F | Page 24 of 27
Data Sheet AD5791
VREFP
2 × VREFN − VREFP to VREFP. This configuration is used to generate
a bipolar output span from a single ended reference input with
1/2 AD8676 VREFN = 0 V. For this mode of operation, the RBUF bit of the
control register must be cleared to Logic 0.
VREFPF VREFPS
VREFP
RFB

R1 6.8kΩ RFB 6.8kΩ 10pF

1/2 AD8676
INV VOUT
20-BIT
DAC VOUT VREFPF VREFPS
AD8675,
ADA4898-1 R1 RFB RFB
A1
AD5791 6.8kΩ 6.8kΩ 10pF
VREFNF VREFNS
INV
VOUT
20-BIT
1/2 AD8676 DAC VOUT
AD8675,

08964-052
ADA4898-1
VREFN AD5791
VREFNF VREFNS
Figure 52. Output Amplifier in Unity Gain with Amplifier Input Bias Current
Compensation
1/2 AD8676
Gain of Two Configuration

08964-053
Figure 53 shows an output amplifier configured for a gain of VREFN = 0V

two. The gain is set by the internal matched 6.8 kΩ resistors, Figure 53. Output Amplifier in Gain of Two Configuration
which are exactly twice the DAC resistance, having the effect of
removing an offset from the input bias current of the external
amplifier. In this configuration, the output spans from

Rev. F | Page 25 of 27
AD5791 Data Sheet

APPLICATIONS INFORMATION
TYPICAL OPERATING CIRCUIT

08964-054
Figure 54. Typical Operating Circuit

Figure 54 shows a typical operating circuit for the AD5791 must be used on the reference inputs. Because the output
using an AD8676 for reference buffers and an AD8675 as an impedance of the AD5791 is 3.4 kΩ, an output buffer is
output buffer. To meet the specified linearity, force sense buffers required for driving low resistive, high capacitance loads.

Rev. F | Page 26 of 27
Data Sheet AD5791

OUTLINE DIMENSIONS
6.60
6.50
6.40

20 11

4.50
4.40
4.30
6.40 BSC
1 10

PIN 1
0.65
BSC
0.15 1.20 MAX 0.20
0.05 0.09 0.75
8° 0.60
0.30
0° 0.45
COPLANARITY 0.19 SEATING
0.10 PLANE

COMPLIANT TO JEDEC STANDARDS MO-153-AC

Figure 55. 20-Lead Thin Shrink Small Outline Package [TSSOP]

(RU-20)
Dimensions shown in millimeters

ORDERING GUIDE
Model 1 Temperature Range INL Package Description Package Option
AD5791BRUZ −40°C to +125°C ±1.5 LSB 20-Lead TSSOP RU-20
AD5791ARUZ −40°C to +125°C ±4 LSB 20-Lead TSSOP RU-20
EVAL-AD5791SDZ Evaluation Board
1
Z = RoHS Compliant Part.

©2010–2020 Analog Devices, Inc. All rights reserved. Trademarks and

registered trademarks are the property of their respective owners.
D08964-0-1/20(F)

Rev. F | Page 27 of 27