AD4696BCPZ [ADI]
16-Bit, 16-Channel, 500 kSPS/1 MSPS, Easy Drive Multiplexed SAR ADCs;型号: | AD4696BCPZ |
厂家: | ADI |
描述: | 16-Bit, 16-Channel, 500 kSPS/1 MSPS, Easy Drive Multiplexed SAR ADCs |
文件: | 总96页 (文件大小:1310K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
16-Bit, 16-Channel, 500 kSPS/1 MSPS,
Easy Drive Multiplexed SAR ADCs
AD4695/AD4696
Data Sheet
FEATURES
GENERAL DESCRIPTION
Easy Drive
The AD4695/AD4696 are compact, high accuracy, low power,
Reduced analog input and reference drive requirements
Overvoltage clamp input current protection up to 5 mA on
each analog input
Long acquisition phase, ≥71.5% (715 ns/1000 ns)of cycle
time at 1 MSPS
High performance
Sample rate: 500 kSPS (AD4695) or 1 MSPS (AD4696)
INL: 1 LSB maximum
Guaranteed 16-bit, no missing codes
SINAD: 93 dB typical, VREF = 5 V and fIN = 1 kHz
Oversampled dynamic range: 111.2 dB, OSR = 64
Small footprint, high channel density
32-lead 5 mm × 5 mm LFCSP
16-channel, 16-bit, 500 kSPS/1 MSPS, multiplexed input
precision, successive approximation register (SAR) analog-to-
digital converters (ADCs) with Easy Drive features and
extensive digital functionality.
The AD4695/AD4696 are optimal for use in space constrained,
multichannel, precision data acquisition systems and monitoring
circuits. The AD4695/AD4696 feature a true 16-bit SAR ADC core
with no missing codes, a 16-channel, low crosstalk multiplexer, a
flexible channel sequencer, overvoltage protection clamp
circuits on each analog input, on-chip oversampling and
decimation, threshold detection and alert indicators, and an
autonomous conversion (autocycle) mode.
The AD4695/AD4696 Easy Drive features relax the drive
requirements of the analog front end (AFE) and reference
circuitry. Analog input high-Z mode and reference input
high-Z mode simplify system designs, reduce component
count, and increase channel density by removing the need for
dedicated high speed ADC drivers and reference buffers.
Easy Drive features support system level designs with
fewer components
Enhanced digital functionality
First conversion accurate, no latency or pipeline delay
Fast conversion time and dual-SDO mode allow low
SPI clock rates
Customizable channel sequencer
On-chip oversampling and decimation
Threshold detection alerts
Input overvoltage protection clamps on each analog input
protect the AD4695/AD4696 from overvoltage events and
prevent overvoltage events on one channel from degrading
performance on other channels (see Figure 26).
Offset and gain correction
Autonomous conversion (autocycle) mode
SPI-/QPSI-/MICROWIRE-/DSP-compatible serial interface
Low power
8 mW at fS = 1 MSPS
4 mW at fS = 500 kSPS
4 µW standby power dissipation with the internal
LDO disabled
Internal LDO enables 3.15 V to 5.5 V, single analog supply
operation
Advanced digital functionality makes the AD4695/AD4696
compatible with a variety of low power digital hosts. The low
serial peripheral interface (SPI) clock rate requirements, on-chip
customizable channel sequencers, and oversampling and
decimation reduce the burden on the digital host system.
Autocycle mode and threshold detection features enable low
power, interrupt driven firmware design by performing
conversions autonomously and generating alerts based on
channel specific threshold limits.
1.14 V to 1.98 V logic interface
Wide operating temperature range: −40°C to +125°C
The AD4695/AD4696 are available in a 5 mm × 5 mm 32-lead
lead frame chip scale package (LFCSP) with operation specified
from −40°C to +125°C.
APPLICATIONS
Photodiode monitoring
Medical instrumentation
Vital signs monitoring
Electronic test and measurement
Automated test equipment
Instrumentation and process control
Battery-powered equipment
Rev. 0
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 license is granted by implication or otherwise under any patent or patent rights of Analog
Devices. Trademarks and registeredtrademarks are the property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700
Technical Support
©2020 Analog Devices, Inc. All rights reserved.
www.analog.com
AD4695/AD4696
Data Sheet
TABLE OF CONTENTS
Features.............................................................................................. 1
Busy Indicator............................................................................. 36
Channel Sequencing Modes ..................................................... 36
Digital Interface.............................................................................. 41
Register Configuration Mode................................................... 41
Conversion Mode....................................................................... 48
Autocycle Mode ......................................................................... 54
General-Purpose Pin ................................................................. 56
Device Reset................................................................................ 56
Applications Information ............................................................. 60
Analog Front-End Design......................................................... 61
Analog Input Overvoltage Protection..................................... 64
Reference Circuitry Design....................................................... 65
Reference Circuit Design for Driving REF Input.................. 65
Converting Between Codes and Volts .................................... 67
Oversampling for Noise Reduction......................................... 67
Digital Interface Operation....................................................... 67
Device Configuration Recommendations .............................. 74
Effective Channel Sample Rate ................................................ 75
Layout Guidelines ...................................................................... 77
Evaluating AD4695/AD4696 Performance............................ 77
Register Information...................................................................... 78
Register Overview ...................................................................... 78
Register Details........................................................................... 79
Outline Dimensions....................................................................... 96
Ordering Guide .......................................................................... 96
Applications ...................................................................................... 1
General Description......................................................................... 1
Revision History ............................................................................... 2
Functional Block Diagram .............................................................. 3
Specifications .................................................................................... 4
Timing Specifications .................................................................. 8
Absolute Maximum Ratings ......................................................... 10
Thermal Resistance.................................................................... 10
Electrostatic Discharge (ESD) Ratings.................................... 10
ESD Caution................................................................................ 10
Pin Configuration and Function Descriptions .......................... 11
Typical Performance Characteristics........................................... 13
Terminology.................................................................................... 23
Theory of Operation ...................................................................... 24
Overview...................................................................................... 24
Converter Operation.................................................................. 24
Transfer Function ...................................................................... 26
Analog Inputs ............................................................................. 27
Input Overvoltage Protection Clamps .................................... 31
Temperature Sensor................................................................... 32
Voltage Reference Input............................................................ 32
Power Supplies............................................................................ 32
Oversampling and Decimation ................................................ 33
Offset and Gain Correction ...................................................... 34
Threshold Detection and Alert Indicators ............................. 34
REVISION HISTORY
12/2020—Revision 0: Initial Version
Rev. 0 | Page 2 of 96
Data Sheet
AD4695/AD4696
FUNCTIONAL BLOCK DIAGRAM
LDO_IN
AVDD
VDD
REF
+1.8V
INTERNAL
LDO
REF INPUT
HIGH-Z MODE
IN0
16-BIT
SAR ADC
OVERVOLTAGE
PROTECTION
CLAMPS
ANALOG
INPUT
INPUT
MUX
HIGH-Z MODE
VIO
CNV
CS
IN15
ON-CHIP OSC,
CONTROL LOGIC
USER REGISTERS,
SEQUENCING LOGIC,
AND
SCK
SDI
COM
TEMPERATURE
SENSOR
DIGITAL DATA
PROCESSING
SDO
BSY_ALT_GP0
RESET
AD4695/AD4696
REFGND
AGND
IOGND
Figure 1.
Rev. 0 | Page 3 of 96
AD4695/AD4696
SPECIFICATIONS
Data Sheet
AVDD = 3.15 V to 5.5 V, LDO_IN = 2.4 V to 5.5 V with internal low dropout (LDO) enabled, LDO_IN = AGND with internal LDO
disabled, VDD = 1.71 V to 1.89 V with internal LDO disabled, VIO = 1.14 V to 1.98 V, AGND = REFGND = IOGND = 0 V, reference
voltage (VREF) = 2.4 V to 5.1 V, REF = VREF, sample rate (fS) = 1 MSPS for the AD4696, fS = 500 kSPS for the AD4695, input frequency
(fIN) = 1 kHz, digital output load capacitance = 20 pF, autocycle mode disabled, analog input high-Z mode enabled, reference input high-Z mode
enabled, busy indicator and alert indicator not enabled on BSY_ALT_GP0, no active overvoltage protection clamps, and TA = −40°C to
+125°C, unless otherwise noted.
Table 1.
Parameter
Test Conditions/Comments
Min
Typ
Max
Unit
RESOLUTION
16
Bits
ANALOG INPUT1, 2
Input Voltage Range
Positive ADC input voltage (IN+) – negative
ADC input voltage (IN−)
Unipolar Mode
0
+VREF
+VREF/2
V
V
Pseudobipolar Mode
Operating Input Voltage
IN+ − REFGND
−VREF/2
IN− = REFGND
0
−0.1
+VREF
VREF + 0.1
V
V
IN− = COM, odd numbered input
IN− = COM, odd numbered input
Unipolar mode
IN− − REFGND
−0.1
VREF/2 –
0.1
VREF + 0.1
VREF/2 + 0.1
V
V
Pseudobipolar mode
VREF/2
Common-Mode Rejection Ratio (CMRR)
Analog Input Leakage Current3
SAMPLING DYNAMICS
Sample Rate
fIN = 250 kHz, IN− = COM, odd numbered input
Autocycle mode disabled
69.5
10
dB
nA
AD4695
AD4696
500
1
kSPS
MSPS
Autocycle Sample Period
Autocycle mode enabled
AC_CYC = 0x0
AC_CYC = 0x1
AC_CYC = 0x2
AC_CYC = 0x3
AC_CYC = 0x4
AC_CYC = 0x5
AC_CYC = 0x6
AC_CYC = 0x7
8.5
17
10
20
11.5
23
μs
μs
34
40
46
μs
68
80
92
μs
85
100
200
400
800
2
115
230
460
920
μs
μs
μs
μs
170
340
680
Aperture Delay
ns
Aperture Jitter
0.5
ps rms
DC ACCURACY
No Missing Codes
16
−1
−0.6
Bits
LSB
LSB
LSB
rms
Integral Nonlinearity Error (INL)
Differential Nonlinearity Error (DNL)
Transition Noise
VREF = 5 V, oversampling ratio (OSR) = 1
VREF = 5 V, OSR = 1
VREF = 5 V
0.4
0.3
0.5
+1
+0.6
Offset Error4
VREF = 5 V, TA = 25°C
0.03
mV
VREF = 5 V, TA = −40°C to +125°C
VREF = 5 V, TA = 25°C
VREF = 5 V, TA = −40°C to +125°C
VREF = 5 V, TA = 25°C
VREF = 5 V, TA = −40°C to +125°C
VREF = 5 V, TA = 25°C
VREF = 5 V, TA = −40°C to +125°C
−0.43
+0.43
mV
mV
mV
%FS5
%FS
%FS
%FS
Offset Error Match4
Gain Error4
0.025
0.001
0.002
−0.23
+0.23
−0.025
−0.012
+0.025
+0.012
Gain Error Match4
Rev. 0 | Page 4 of 96
Data Sheet
AD4695/AD4696
Parameter
Test Conditions/Comments
Min
Typ
Max
Unit
AC PERFORMANCE
Dynamic Range
VREF = 5 V
OSR = 1
OSR = 4
OSR = 16
OSR = 64
OSR = 1
OSR = 4
OSR = 16
OSR = 64
Bandwidth = 0.1 Hz to 10 Hz
VREF = 5 V, fIN = 1 kHz
VREF = 4.096 V, fIN = 1 kHz
VREF = 2.5 V, fIN = 1 kHz
VREF = 5 V, fIN = 1 kHz
VREF = 4.096 V, fIN = 1 kHz
VREF = 2.5 V, fIN = 1 kHz
VREF = 5 V, fIN = 1 kHz
VREF = 4.096 V, fIN = 1 kHz
VREF = 2.5 V, fIN = 1 kHz
VREF = 5 V
93.4
99.3
105.3
111.2
37.8
19.2
9.6
4.9
5
93
91.3
87
−117
−117.5
−119
93
91.3
87
121
−123
−100
−110
11.7
dB
dB
dB
dB
μV rms
μV rms
μV rms
μV rms
μV p-p
dB
dB
dB
dB
dB
dB
dB
dB
dB
Input RMS Noise
1/f Noise
Signal-to-Noise Ratio (SNR)
91.25
Total Harmonic Distortion (THD)
Signal-to-Noise-and-Distortion (SINAD)
Spurious-Free Dynamic Range (SFDR)
Channel to Channel Isolation
Channel to Channel Memory
dB
dB
dB
dB
fIN = 100 kHz
fIN = 100 kHz, fS = 1 MSPS
fIN = 100 kHz, fS = 500 kSPS
−3 dB Input Bandwidth
REFERENCE
MHz
VREF Range
2.4
AVDD +
0.25
V
REF Leakage Current
VREF = 5 V
No active overvoltage protection clamps
All clamps active, overvoltage reduced,
current mode disabled
165
375
nA
μA
All clamps active, overvoltage reduced,
current mode enabled
8
μA
REF Average Input Current
VREF = AVDD = 5 V
Reference High-Z Mode Disabled
fS = 10 kSPS, unipolar mode
fS = 500 kSPS, unipolar mode
fS = 1 MSPS, unipolar mode
fS = 10 kSPS, pseudobipolar mode
fS = 500 kSPS, pseudobipolar mode
fS = 1 MSPS, pseudobipolar mode
fS = 10 kSPS, unipolar mode
fS = 500 kSPS, unipolar mode
fS = 1 MSPS, unipolar mode
fS = 10 kSPS, pseudobipolar mode
fS = 500 kSPS, pseudobipolar mode
fS = 1 MSPS, pseudobipolar mode
3.3
160
320
4.0
195
390
0.3
6
μA
μA
µA
μA
μA
µA
μA
μA
μA
μA
μA
µA
Reference High-Z Mode Enabled
12
0.4
11
22
TEMPERATURE SENSOR
Temperature Sensor Voltage
TA = 25°C
TA = 0°C
TA = −40°C to +125°C
680
725
−1.8
mV
mV
mV/°C
Temperature Sensitivity
Rev. 0 | Page 5 of 96
AD4695/AD4696
Data Sheet
Parameter
Test Conditions/Comments
Min
Typ
Max
Unit
OVERVOLTAGE CLAMP
External Series Resistance (REXT
6
)
For stable clamp operation
Overvoltage reduced current mode disabled
Overvoltage reduced current mode enabled
For stable clamp operation
2000
1000
Ω
Ω
pF
mA
V
V
V
ns
ns
6
External Series Capacitance (CEXT
Clamp Input Current
Clamp Activation Voltage
Clamp Deactivation Voltage
Input Clamping Voltage
Activation Time
)
500
For each active clamp
5
VREF + 0.55
VREF + 0.1
Clamp current = 5 mA
VREF + 0.2
50
100
Deactivation Time
DIGITAL INPUTS
Logic Levels
Input Low Voltage (VIL)
Input High Voltage (VIH)
Input Current (IL)
−0.3
0.7 × VIO
−1
+0.3 × VIO
3.6
+1
V
V
µA
pF
Input Pin Capacitance
DIGITAL OUTPUTS
5
Conversion Mode Data Format
Unipolar mode
Straight binary
Pseudo bipolar mode
Twos complement
Logic Levels
Output Low Voltage (VOL)
Output High Voltage (VOH)
POWER REQUIREMENTS
AVDD to AGND
Digital output current = +500 µA
Digital output current = −500 µA
0.4
V
V
VIO − 0.3
3.15
2.4
5.5
5.5
V
V
V
V
V
LDO_IN to AGND
Internal LDO enabled
Internal LDO disabled
Internal LDO disabled
0
1.8
VDD to AGND
VIO to IOGND
POWER SUPPLY CURRENT7
Standby Current
AVDD
1.71
1.14
1.89
1.98
AVDD = 5 V
160
nA
LDO_IN
LDO_IN = 5 V
Internal LDO enabled
Internal LDO disabled
VDD = 1.8 V, internal LDO disabled
VIO = 1.8 V
9
µA
µA
µA
nA
0.3
1.5
250
VDD
VIO
AVDD Current (Conversion Mode)
AVDD = 5 V
Reference High-Z Mode Disabled,
Analog Input High-Z Mode Disabled
fS = 10 kSPS
680
nA
fS = 500 kSPS
fS = 1 MSPS
fS = 10 kSPS
26
52
13
µA
µA
µA
Reference High-Z Mode Enabled,
Analog Input High-Z Mode Enabled
fS = 500 kSPS
fS = 1 MSPS
0.64
1.28
0.73
1.46
mA
mA
LDO_IN Current (Conversion Mode)
Reference High-Z Mode Disabled,
Analog Input High-Z Mode Disabled
LDO_IN = 5 V, internal LDO enabled
fS = 10 kSPS
52
µA
fS = 500 kSPS
fS = 1 MSPS
fS = 10 kSPS
2
4
64
mA
mA
µA
Reference High-Z Mode Enabled,
Analog Input High-Z Mode Enabled
Rev. 0 | Page 6 of 96
Data Sheet
AD4695/AD4696
Parameter
Test Conditions/Comments
fS = 500 kSPS
fS = 1 MSPS
Min
Typ
2.6
5.2
Max
3.3
6.6
Unit
mA
mA
VDD Current (Conversion Mode)
Reference High-Z Mode Disabled,
Analog Input High-Z Mode Disabled
VDD = 1.8 V, internal LDO disabled
fS = 10 kSPS
42
µA
fS = 500 kSPS
fS = 1 MSPS
fS = 10 kSPS
2
4
53
mA
mA
µA
Reference High-Z Mode Enabled,
Analog Input High-Z Mode Enabled
fS = 500 kSPS
fS = 1 MSPS
2
5
3.2
6.4
mA
mA
VIO Dynamic Current
Register Configuration Mode
Conversion Mode
VIO = 1.8 V
Streaming mode, SCK frequency (fSCK) = 50 MHz
Status bits enabled
fS = 10 kSPS
fS = 500 kSPS
fS = 1 MSPS
125
µA
3.5
162
325
µA
µA
µA
360
POWER DISSIPATION7
AVDD = 5 V, VIO = 1.8 V
Standby Power Dissipation
Internal LDO Disabled
Internal LDO Enabled
VDD = 1.8 V
LDO_IN = 5 V
4
46
µW
µW
Power Dissipation, Internal LDO Disabled
Reference High-Z Mode Disabled,
Analog Input High-Z Mode Disabled
LDO_IN = AGND, VDD = 1.8 V
fS = 10 kSPS
85
µW
fS = 500 kSPS
fS = 1 MSPS
fS = 10 kSPS
4
8
170
mW
mW
µW
Reference High-Z Mode Enabled,
Analog Input High-Z Mode Enabled
fS = 500 kSPS
fS = 1 MSPS
LDO_IN = 5 V
fS = 10 kSPS
8
16
9.8
19.5
mW
mW
Power Dissipation, Internal LDO Enabled
Reference High-Z Mode Disabled,
Analog Input High-Z Mode Disabled
270
µW
fS = 500 kSPS
fS = 1 MSPS
fS = 10 kSPS
10.5
21
395
mW
mW
µW
Reference High-Z Mode Enabled,
Analog Input High-Z Mode Enabled
fS = 500 kSPS
fS = 1 MSPS
16.5
33
20.5
41.0
mW
mW
Autocycle Mode Power Dissipation
LDO_IN = 5 V, internal LDO enabled,
autocycle mode enabled
AC_CYC = 0x0
AC_CYC = 0x7
2.3
0.2
mW
mW
TEMPERATURE RANGE
Specified Performance
TMIN to TMAX
−40
+125
°C
1 See the Channel Configuration Options section for a detailed description of unipolar mode, pseudo bipolar mode, and the channel pin assignment options.
2 IN+ and IN− represent the analog inputs connected to the positive and negative inputs of the AD4695/AD4696 ADC core via the internal multiplexer (see the
Multiplexer section and Channel Configuration Options section).
3 The analog input leakage current specification refers to the input current of the analog input pins during periods when the ADC is not performing conversions and the
analog input voltage is already settled.
4 Offset error and gain error specifications are taken with the offset and gain correction registers set to the default values, which correspond to no offset or gain
correction. See the Offset and Gain Correction section for more information.
5 %FS is the percentage of the ADC full-scale (see the Transfer Function section for a definition of full scale).
6 REXT and CEXT refer to the resistor and capacitor, respectively, that make up the recommended external RC filters at the analog inputs (see the External RC Filter section).
7 For the power supply current and power dissipation specifications where analog input high-Z mode is enabled, analog input high-Z mode is set to be enabled for all
channels. The power consumption scales with the percentage of conversions performed with analog input high-Z mode enabled.
Rev. 0 | Page 7 of 96
AD4695/AD4696
Data Sheet
TIMING SPECIFICATIONS
AVDD= 3.15 V to 5.5 V, LDO_IN = 2.4 V to 5.5 V with internal LDO enabled, LDO_IN = AGND with internal LDO disabled, VDD =
1.71 V to 1.89 V with internal LDO disabled, VIO = 1.14 V to 1.98 V, AGND = REFGND = IOGND = 0 V, reference voltage (VREF) = 2.4 V
to 5.1 V, REF = VREF, fS = 1 MSPS for the AD4696, fS = 500 kSPS for the AD4695, digital output load capacitance = 20 pF, autocycle mode
disabled, no active overvoltage protection clamps, and TA = −40°C to +125°C, unless otherwise noted.
Table 2.
Parameter1
Symbol
tCONVERT
tACQ
Min
Typ Max Unit
Conversion Time
Acquisition Time
380 415
ns
Two-Cycle Command Mode, Standard Sequencer, or Advanced Sequencer Enabled
fS = 1 MSPS
fS = 500 kSPS
Single-Cycle Command Mode2 Enabled
715
1715
ns
ns
CNV Period (Time Between Conversions)
tCYC
fS = 1 MSPS, Autocycle Mode Disabled
fS = 500 kSPS, Autocycle Mode Disabled
1000
2000
ns
ns
Autocycle Mode Enabled
AC_CYC = 0x0
AC_CYC = 0x1
AC_CYC = 0x2
AC_CYC = 0x3
AC_CYC = 0x4
AC_CYC = 0x5
AC_CYC = 0x6
AC_CYC = 0x7
8.5
17
34
68
85
170
340
680
10
80
5
10
20
40
80
11.5 μs
23
46
92
μs
μs
μs
μs
μs
μs
μs
ns
ns
ns
ns
ns
100 115
200 230
400 460
800 920
CNV High Time
CNV Low Time
CS High Time
tCNVH
tCNVL
tCSBH
tEN
CS Low to Digital Interface Ready Delay
15
15
CS High to SDO High Impedance Delay
tCSBDIS
tSCK
SCK Period
Register Configuration Mode
Conversion Mode
40
12.5
ns
ns
SCK Low Time
tSCKL
Register Configuration Mode
Conversion Mode
16
5
ns
ns
SCK High Time
tSCKH
Register Configuration Mode
Conversion Mode
SDI Data Setup Time Prior to SCK Rising Edge
SDI Data Hold Time After SCK Rising Edge
SCK Falling Edge to Data Remains Valid Delay
SCK Falling Edge to Data Valid Delay
Last SCK Edge to CNV Rising Edge Delay
Last SCK Rising Edge to CS Rising Edge Delay
16
5
2
2
1.5
ns
ns
ns
ns
ns
tSSDI
tHSDI
tHSDO
tDSDO
tSCKCNV
tSCKCSB
tCNVBSY
tCNVALT
tACBSY
10.5 ns
80
1
ns
ns
CNV Rising Edge to Busy Indicator Rising Edge (Busy Indicator Enabled on General-Purpose Pin)
CNV Rising Edge to Alert Indicator Transition (Alert Indicator Enabled on General-Purpose Pin)
Busy Indicator Low Time, Autocycle Mode Enabled (Busy Indicator Enabled on General-Purpose Pin)
20
425
ns
ns
AC_CYC = 0x0
AC_CYC = 0x1
AC_CYC = 0x2
AC_CYC = 0x3
AC_CYC = 0x4
8
µs
µs
µs
µs
µs
16.5
33.5
67.5
84.5
Rev. 0 | Page 8 of 96
Data Sheet
AD4695/AD4696
Parameter1
Symbol
Min
Typ Max Unit
AC_CYC = 0x5
AC_CYC = 0x6
AC_CYC = 0x7
Register Configuration Mode Setup Time
RESET Low Time
169
339
679
20
µs
µs
µs
ns
ns
µs
µs
ms
ms
ms
ms
ms
ms
tREGCONFIG
tRESETL
10
Hardware Reset Delay (VDD Always Supplied)
Software Reset Delay
VDD Power-On Reset Delay
VIO Power-On Reset Delay (VDD Supplied Externally)
LDO_IN Power-On Reset Delay
VIO Power-On Reset Delay (VDD Supplied by Internal LDO)
LDO Wake-Up Command Power-On Reset Delay
Hardware Reset Delay (Internal LDO Disabled)
tHWR_DELAY 310
tSWR_DELAY 310
tPOR_VDD
tPOR_VIO1
tPOR_LDO
2
1.3
3.2
3
3
3
tPOR_VIO2
tWAKEUP_SW
tWAKEUP_HW
1 For all specifications, the relative voltages for the AVDD and REF inputs follow the operating conditions specified in the reference and power requirements sections of Table 1.
2 The acquisition time for single-cycle command mode depends on the sample rate and SCK frequency (see the Single-Cycle Command Mode section).
Rev. 0 | Page 9 of 96
AD4695/AD4696
Data Sheet
ABSOLUTE MAXIMUM RATINGS
Table 3.
THERMAL RESISTANCE
θJA is specified for worst case conditions and is the natural
convection junction to ambient thermal resistance measured in
a one cubic foot sealed enclosure. θJC is the junction to case
thermal resistance.
Parameter
Rating
Analog Inputs
INn1, COM to REFGND
Reference Inputs
REF to AGND, REFGND, IOGND
Supply Inputs
−0.3 V to REF + 0.3 V
−0.3 V to +6 V
Thermal resistance values specified in Table 4 were calculated
based on JEDEC specifications and must be used in compliance
with JESD51-12. The worst case junction temperature is reported.
AVDD, LDO_IN to AGND, REFGND, IOGND −0.3 V to +6 V
VDD, VIO to AGND, REFGND, IOGND
AVDD to LDO_IN
AVDD, LDO_IN to REF
VDD, VIO to AVDD, LDO_IN, REF
VDD to VIO
−0.3 V to +2.1 V
−6.3 V to +6.3 V
−6.3 V to +6.3 V
−6.3 V to +2.4 V
−2.4 V to +2.4 V
θJA is highly dependent on the application and board layout. In
applications where high maximum power dissipation exists,
close attention to thermal board design is required. The θJA
value can vary depending on printed circuit board (PCB) material,
layout, and environmental conditions.
Ground
Table 4. Thermal Resistance
AGND, IOGND to REFGND
AGND to IOGND
Digital Inputs2 to IOGND
Digital Outputs2 to IOGND
Storage Temperature Range
Junction Temperature
−0.3 V to +0.3 V
−0.3 V to +0.3 V
−0.3 V to +6 V
−0.3 V to VIO + 0.3 V
−65°C to +150°C
150°C
Package Type
θJA
40.21
θJC
17.52
Unit
CP-32-7
°C/W
1 Simulated values are based on the JEDEC 2S2P thermal test board with nine
thermal vias in a JEDEC natural convection environment. See JEDEC JESD51.
2 Simulated values are measured to the package top surface with a cold plate
attached directly to the package top surface.
Lead Temperature Soldering
260°C reflow, as per
JEDEC J-STD-020
ELECTROSTATIC DISCHARGE (ESD) RATINGS
The following ESD information is provided for handling of
ESD-sensitive devices in an ESD protected area only.
1 INn refers to the analog inputs, Pin IN0 through Pin IN15.
2 See the Pin Configuration and Function Descriptions section for a list of the
digital input and digital output pins.
Human body model (HBM) per ANSI/ESDA/JEDEC JS-001.
Stresses at or above those listed under Absolute Maximum
Ratings may cause permanent damage to the product. This is a
stress rating only; functional operation of the product at these
or any other conditions above those indicated in the operational
section of this specification is not implied. Operation beyond
the maximum operating conditions for extended periods may
affect product reliability.
Field induced charged device model (FICDM) per
ANSI/ESDA/JEDEC JS-002.
ESD Ratings for AD4695/AD4696
Table 5. AD4695/AD4696, 32-Lead LFCSP
ESD Model
Withstand Threshold (kV)
Class
3A
HBM
4
FICDM
1.25
C3
ESD CAUTION
Rev. 0 | Page 10 of 96
Data Sheet
AD4695/AD4696
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
IN5 1
24 CNV
23
2
IN6
IN7 3
SDI
22 SCK
AD4695/
AD4696
COM
IN8
IN9
SDO
4
5
6
21
20
19
BSY_ALT_GP0
IOGND
TOP VIEW
(Not to Scale)
IN10 7
IN11 8
18 VIO
17
RESET
NOTES
1. EXPOSED PAD. THE EXPOSED PAD IS NOT
CONNECTED INTERNALLY. FOR INCREASED
RELIABILITY OF THE SOLDER JOINTS, IT IS
RECOMMENDED THAT THE PAD BE SOLDERED
TO THE SYSTEM GROUND PLANE.
Figure 2. Pin Configuration
Table 6. Pin Function Descriptions
Pin
No.
Mnemonic.
IN5
IN6
IN7
COM
Type1 Description
1
2
3
4
AI
AI
AI
AI
Analog Input 5.
Analog Input 6.
Analog Input 7.
Common Channel Input. IN0 to IN15 can be paired with COM for the ADC core to sample the differential voltage
between them. COM is nominally tied to signal ground (unipolar mode) or VREF/2 (pseudobipolar mode).
See the Channel Configuration Options section for a detailed description on pairing inputs, unipolar
mode, and pseudobipolar mode.
5
6
7
8
IN8
IN9
AI
AI
AI
AI
AI
AI
AI
AI
P
Analog Input 8.
Analog Input 9.
Analog Input 10.
Analog Input 11.
Analog Input 12.
Analog Input 13.
Analog Input 14.
Analog Input 15.
IN10
IN11
IN12
IN13
IN14
IN15
AGND
AVDD
LDO_IN
9
10
11
12
13
14
15
Analog Supply Ground. AVDD, LDO_IN, and VDD are referenced to AGND.
Analog Power Supply. AVDD is nominally 3.15 V to 5.5 V. Decouple AVDD to AGND with a local 100 nF capacitor.
Internal LDO Input. LDO_IN is nominally 2.4 V to 5.5 V when the internal LDO is enabled. Decouple LDO_IN
to AGND with a local 100 nF capacitor. If powering VDD with an external 1.8 V rail, tie LDO_IN to AGND.
See the Internal LDO section for more information.
P
P
16
VDD
P
ADC Core Power Supply. VDD is nominally 1.8 V. When powering VDD from the internal LDO, leave VDD
floating. When powering VDD from an external rail, disable the internal LDO. Decouple VDD to AGND with
a local 100 nF capacitor.
17
18
RESET
VIO
DI
P
Hardware Reset Input. Drive RESET low to perform a hardware reset of the device and reset the register
states to the default values (see the Device Reset section).
Input/Output Interface Digital Power. VIO is nominally the same supply as the host interface (for example,
1.2 V to 1.8 V). Decouple VIO to IOGND with a local 100 nF capacitor.
19
20
IOGND
P
Input/Output Interface Digital Supply Ground. VIO is referenced to IOGND.
BSY_ALT_GP0 DI/DO General-Purpose Pin 0. BSY_ALT_GP0 can be configured to function as a general-purpose input/output
(GPIO), the threshold detection alert indicator, the busy indicator, or a second serial data output (see the
General-Purpose Pin section).
21
SDO
DO
Serial Data Output. When the device is configured in register configuration mode, SDO is used to read the
configuration register data during SPI read transactions. When the device is configured in conversion mode,
SDO is used to read the conversion results. Data output is synchronized to the falling edge of SCK.
Rev. 0 | Page 11 of 96
AD4695/AD4696
Data Sheet
Pin
No.
Mnemonic.
Type1 Description
22
SCK
DI
Serial Data Clock Input. SCK is used to clock out data on SDO and clock in data on SDI while the device is
configured in either register configuration mode or conversion mode.
23
24
25
SDI
CNV
CS
DI
Serial Data Input. When the device is configured in register configuration mode, SDI is used to perform SPI
read and write transactions to access the configuration registers. In conversion mode, SDI receives 5-bit
commands from the digital host, as shown in Table 16.
Convert Input. When the device is configured in conversion mode, a rising edge on CNV initiates a
conversion of the selected analog input. The AD4695/AD4696 can interface to a 4-wire SPI by tying CNV to
CS. See the Digital Interface Operation section for more information.
DI
DI
Chip Select Input. When configured in register configuration mode, CS frames SPI read and write transactions
that accesses the configuration registers. When the device is configured in conversion mode, CS can either be
held low throughout the entire conversion or used to frame SPI transactions that read back conversion
results. The AD4695/AD4696 can interface to a 4-wire SPI by tying CNV to CS. See the Digital Interface
Operation section for more information.
26
27
REFGND
REF
P
Reference Ground. REF is referenced to REGND. IN0 to IN15 can be paired with REFGND to the ADC core to
sample the differential voltage between them. See the Channel Configuration Options section for a
detailed description on pairing inputs.
Reference Input. VREF must be provided by an external precision reference voltage between 2.4 V and
5.1 V. The REF pin must be decoupled with a minimum 1 μF capacitor for optimal operation. See the
Voltage Reference Input section for more information.
AI
28
29
30
31
32
33
IN0
IN1
IN2
IN3
IN4
EPAD
AI
AI
AI
AI
AI
NC
Analog Input 0.
Analog Input 1.
Analog Input 2.
Analog Input 3.
Analog Input 4.
Exposed Pad. The exposed pad is not connected internally. For increased reliability of the solder joints, it is
recommended that the pad be soldered to the system ground plane.
1 AI is analog input, P is power, DI is digital input, DO is digital output, and NC is no internal connection.
Rev. 0 | Page 12 of 96
Data Sheet
AD4695/AD4696
TYPICAL PERFORMANCE CHARACTERISTICS
AVDD = LDO_IN = 5 V, VIO = 1.8 V, VREF = 5 V, fSCK = 50 MHz, unipolar mode, analog input high-Z mode enabled, reference input
high-Z mode enabled, internal LDO enabled, fS = 1 MSPS for the AD4696, fS = 500 kSPS for the AD4695, no active clamps, autocycle
mode disabled, OSR = 1, and TA = 25°C, unless otherwise specified.
0.5
0.5
+125°C
+25°C
–40°C
+125°C
+25°C
–40°C
0.4
0.4
0.3
0.3
0.2
0.2
0.1
0.1
0
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.1
–0.2
–0.3
–0.4
–0.5
0
0
0
8192 16384 24576 32768 40960 49152 57344 65536
CODE
0
0
0
8192 16384 24576 32768 40960 49152 57344 65536
CODE
Figure 3. INL vs. Code, VREF = 5 V
Figure 6. DNL vs. Code, VREF = 5 V
0.5
0.4
0.5
0.4
+125°C
+25°C
–40°C
+125°C
+25°C
–40°C
0.3
0.3
0.2
0.2
0.1
0.1
0
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.1
–0.2
–0.3
–0.4
–0.5
8192 16384 24576 32768 40960 49152 57344 65536
CODE
8192 16384 24576 32768 40960 49152 57344 65536
CODE
Figure 4. INL vs. Code, VREF = 4.096 V
Figure 7. DNL vs. Code, VREF = 4.096 V
0.5
0.4
0.5
0.4
+125°C
+25°C
–40°C
+125°C
+25°C
–40°C
0.3
0.3
0.2
0.2
0.1
0.1
0
0
–0.1
–0.2
–0.3
–0.4
–0.5
–0.1
–0.2
–0.3
–0.4
–0.5
8192 16384 24576 32768 40960 49152 57344 65536
CODE
8192 16384 24576 32768 40960 49152 57344 65536
CODE
Figure 5. INL vs. Code, VREF = 2.5 V
Figure 8. DNL vs. Code, VREF = 2.5 V
Rev. 0 | Page 13 of 96
AD4695/AD4696
Data Sheet
900000
800000
700000
600000
500000
400000
300000
200000
100000
0
600000
500000
400000
300000
200000
100000
0
V
V
= 5.0V
= 2.5V
V
V
= 5.0V
= 2.5V
REF
REF
REF
REF
32764 32765 32766 32767 32768 32769 32770 32771 32772
ADC CODE
32764 32765 32766 32767 32768 32769 32770 32771
ADC CODE
Figure 9. Histogram of a DC Input at Code Center, OSR = 1
Figure 12. Histogram of a DC Input at Code Transition, OSR = 1
50000
45000
40000
35000
30000
25000
20000
15000
10000
5000
50000
V
V
= 5.0V
= 2.5V
V
V
= 5.0V
= 2.5V
REF
REF
REF
REF
45000
40000
35000
30000
25000
20000
15000
10000
5000
0
0
ADC CODE
ADC CODE
Figure 10. Histogram of a DC Input at Code Center, OSR = 64
Figure 13. Histogram of a DC Input at Code Transition, OSR = 64
112
110
108
106
104
102
100
98
674
673
672
671
670
669
668
96
94
92
1
4
16
64
0
1
2
3
4
5
6
7
8
9
10
OSR
TIME (SECONDS)
Figure 11. Dynamic Range vs. OSR
Figure 14. 1/f Noise (0.1 Hz to 10 Hz Bandwidth), 50 kSPS,
2500 Samples Averaged per Reading
Rev. 0 | Page 14 of 96
Data Sheet
AD4695/AD4696
0
0
–20
262144 SAMPLES
= 5V
SNR = 93.4dB
THD = –119.3dB
SINAD = 93.4dB
262144 SAMPLES
V
V
= 2.5V
REF
REF
–20
SNR = 88.8dB
THD = –120.9dB
SINAD = 88.8dB
–40
–40
–60
–60
–80
–80
–100
–140
–180
–100
–140
–180
–200
–200
100
1000
10000
100000
100
1000
10000
100000
FREQUENCY (Hz)
FREQUENCY (Hz)
Figure 15. Fast Fourier Transform (FFT), fIN = 1 kHz, VREF = 5 V,
OSR = 1
Figure 18. FFT, fIN = 1 kHz, VREF = 2.5 V, OSR = 1
0
0
–20
65536 SAMPLES
REF
SNR = 110.3dB
THD = –119.7dB
SINAD = 109.9dB
65536 SAMPLES
V
= 5V
V
= 5V
REF
–20
–40
SNR = 106.1dB
THD = –125.6dB
SINAD = 106.0dB
–40
–60
–60
–80
–80
–100
–140
–180
–200
–100
–140
–180
–200
10
100
FREQUENCY (Hz)
1000
10
100
FREQUENCY (Hz)
1000
Figure 16. FFT, fIN = 1 kHz, VREF = 5 V, OSR = 64
Figure 19. FFT, fIN = 1 kHz, VREF = 2.5 V, OSR = 64
94
–80
5.10
5.05
5.00
4.95
4.90
4.85
4.80
4.75
4.70
4.65
4.60
–85
93
92
91
90
89
88
87
–90
–95
SNR
SINAD
THD
–100
–105
–110
–115
–120
–125
R
= 200Ω, C
= 500pF
EXT
EXT
ANALOG INPUT HIGH-Z MODE DISABLED
ANALOG INPUT HIGH-Z MODE ENABLED
100
1k
10k
100k
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
INPUT FREQUENCY (Hz)
TIME (μs)
Figure 17. SNR, SINAD, and THD vs. Input Frequency
Figure 20. Analog Input Voltage Step with Analog Input High-Z Mode
Disabled and Enabled
Rev. 0 | Page 15 of 96
AD4695/AD4696
Data Sheet
94
15.3
15.2
15.1
15.0
14.9
14.8
14.7
14.6
14.5
14.4
14.3
–113
–114
–115
–116
–117
–118
–119
121
120
119
118
117
116
115
SNR
SINAD
93
ENOB
THD
SFDR
92
91
90
89
88
2.4
2.7
3.0
3.3
3.6
3.9
(V)
4.2
4.5
4.8
5.3
2.4
2.7
3.0
3.3
3.6
3.9
4.2
4.5
4.8
5.1
V
TEMPERATURE (°C)
REF
Figure 24. THD and SFDR vs. Temperature, fIN = 1 kHz
Figure 21. SNR, SINAD, and Effective Number of Bits (ENOB) vs. VREF
fIN = 1 kHz
,
–104
–106
–108
–110
–112
–114
–116
–118
–120
–122
130
125
120
115
110
105
94.0
93.8
93.6
93.4
93.2
93.0
92.8
92.6
92.4
15.26
15.24
15.22
15.20
15.18
15.16
15.14
15.12
15.10
15.08
15.06
THD
SNR
SFDR
SINAD
ENOB
–40
–20
0
20
40
60
80
100
120
–40
–20
0
20
40
60
80
100
120
TEMPERATURE (°C)
TEMPERATURE (°C)
Figure 25. THD and SFDR vs. Temperature, fIN = 1 kHz
Figure 22. SNR, SINAD, and ENOB vs. Temperature, fIN = 1 kHz
93.35
93.30
93.25
93.20
93.15
93.10
93.05
93.00
92.95
92.90
93.35
93.30
SNR (dB)
SNR (dB)
SINAD (dB)
SINAD (dB)
93.25
93.20
93.15
93.10
93.05
93.00
92.95
92.90
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
NUMBER OF ACTIVE CLAMPS
NUMBER OF ACTIVE CLAMPS
Figure 26. SNR, SINAD vs. Number of Active Clamps,
Reduced Current Mode Enabled
Figure 23. SNR, SINAD vs. Number of Active Clamps,
Reduced Current Mode Disabled
Rev. 0 | Page 16 of 96
Data Sheet
AD4695/AD4696
100
0.0020
0.0015
0.0010
0.0005
0
IN0
IN1
IN2
IN3
IN4
IN5
IN6
IN7
IN8
IN0
IN1
IN2
IN3
IN4
IN5
IN6
IN7
IN8
IN9
IN9
80
IN10
IN11
IN12
IN13
IN14
IN15
IN10
IN11
IN12
IN13
IN14
IN15
60
40
20
0
–20
–40
–60
–80
–0.0005
–0.0010
–0.0015
–40
–20
0
20
40
60
80
100
120
–40
–20
0
20
40
60
80
100
120
TEMPERATURE (°C)
TEMPERATURE (°C)
Figure 27. Offset Error vs. Temperature
Figure 30. Gain Error vs. Temperature
30
0.00100
0.00075
0.00050
0.00025
0
20
10
0
–10
–20
–30
–40
–0.00025
–0.00050
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
NUMBER OF ACTIVE CLAMPS
NUMBER OF ACTIVE CLAMPS
Figure 28. Offset Error vs. Number of Active Clamps,
Clamp Current = 5 mA, Reduced Current Mode Disabled
Figure 31. Gain Error vs. Number of Active Clamps,
Clamp Current = 5 mA, Reduced Current Mode Disabled
30
0.00100
20
10
0.00075
0.00050
0.00025
0
0
–10
–20
–30
–40
–0.00025
–0.00050
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
NUMBER OF ACTIVE CLAMPS
NUMBER OF ACTIVE CLAMPS
Figure 29. Offset Error vs. Number of Active Clamps,
Clamp Current = 5 mA, Reduced Current Mode Enabled
Figure 32. Gain Error vs. Number of Active Clamps,
Clamp Current = 5 mA, Reduced Current Mode Enabled
Rev. 0 | Page 17 of 96
AD4695/AD4696
Data Sheet
90
85
80
75
70
65
60
55
50
–100
–105
–110
–115
–120
–125
–130
–135
–140
–145
–150
AGGRESSOR ON 1 CHANNEL
AGGRESSOR ON 15 CHANNELS
100
1000
10000
100000
1000000
1k
10k
100k
INPUT FREQUENCY (Hz)
INPUT FREQUENCY (Hz)
Figure 33. CMRR vs. Input Frequency
Figure 36. Channel to Channel Isolation vs. Input Frequency
110
0.85
0.80
0.75
0.70
0.65
0.60
0.55
0.50
0.45
0.40
100
90
80
70
60
50
40
AVDD
LDO_IN
VDD
VIO
100
1k
10k
100k
1M
–40
–20
0
20
40
60
80
100
120
FREQUENCY (Hz)
TEMPERATURE (°C)
Figure 34. PSRR vs. Frequency
Figure 37. Temperature Sensor Output vs. Temperature
5.4
5.2
5.0
4.8
4.6
4.4
4.2
4.0
6
5
4
3
2
1
0
SOURCE VOLTAGE RISING
SOURCE VOLTAGE FALLING
4
5
6
7
8
9
10
4
5
6
7
8
9
10
ANALOG SOURCE VOLTAGE (V)
ANALOG SOURCE VOLTAGE (V)
Figure 35. Analog Input Voltage vs. Analog Source Voltage,
REXT = 1 kΩ, VREF = 5 V
Figure 38. Analog Input Current vs. Analog Source Voltage,
REXT = 1 kΩ, VREF = 5 V
Rev. 0 | Page 18 of 96
Data Sheet
AD4695/AD4696
0.35
400
350
300
250
200
150
100
50
OVERVOLTAGE REDUCED CURRENT MODE DISABLED
OVERVOLTAGE REDUCED CURRENT MODE ENABLED
REFERENCE INPUT HIGH-Z MODE DISABLED
REFERENCE INPUT HIGH-Z MODE ENABLED
0.30
0.25
0.20
0.15
0.10
0.05
0
0
0
100 200 300 400 500 600 700 800 900 1000
SAMPLE RATE (kSPS)
1
3
5
7
9
11
13
15
17
NUMBER OF ACTIVE CLAMPS
Figure 39. Reference Input Current vs. Sample Rate, VREF = 5 V
Figure 42. Additional Reference Input Current vs. Number of Active
Clamps, Clamp Current = 5 mA, VREF = 5 V
0.35
6
0.30
5
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
LDO_IN
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
0.25
0.20
0.15
0.10
0.05
0
4
3
VDD (C
= 1µF)
VDD
VDD
VDD (C
= 10µF)
2
1
0
–1
–0.5
–40
–20
0
20
40
60
80
100
120
0
0.5
1.0
1.5
2.0
TEMPERATURE (°C)
TIME (ms)
Figure 40. Reference Input Current vs. Temperature, VREF = 5 V,
Frequency of CNV Signal (fCNV) = 1 MSPS
Figure 43. LDO_IN, VDD Voltage vs. Time,
CVDD is the VDD Decoupling Capacitance
0.35
0.30
0.25
0.20
0.15
14
12
10
8
VIO = 1.2V
VIO = 1.8V
6
0.10
4
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
0.05
0
2
0
2.4
2.7
3
3.3
3.6
3.9
4.2
4.5
4.8
5.1
0
50
100
150
200
250
REFERENCE VOLTAGE (V)
DIGITAL OUTPUT LOAD CAPACITANCE (pF)
Figure 41. Reference Input Current vs. Reference Voltage,
fS = 1 MSPS and 500 kSPS
Figure 44. tDSDO vs. Digital Output Load Capacitance
Rev. 0 | Page 19 of 96
AD4695/AD4696
Data Sheet
1.4
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
ANALOG INPUT HIGH-Z MODE,
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED
ANALOG INPUT HIGH-Z MODE,
1.2
1.0
0.8
0.6
0.4
0.2
0
REFERENCE INPUT HIGH-Z MODE ENABLED
REFERENCE INPUT HIGH-Z MODE ENABLED
0
100 200 300 400 500 600 700 800 900 1000
SAMPLE RATE (kSPS)
0
100 200 300 400 500 600 700 800 900 1000
SAMPLE RATE (kSPS)
Figure 45. AVDD Current vs. Sample Rate
Figure 48. LDO_IN Current vs. Sample Rate, Internal LDO Enabled
1.6
6.5
ANALOG INPUT HIGH-Z MODE,
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
3.15
3.65
4.15
4.65
5.15
2.4
2.9
3.4
3.9
4.4
4.9
5.4
AVDD VOLTAGE (V)
LDO_IN VOLTAGE (V)
Figure 46. AVDD Current vs. AVDD Voltage
Figure 49. LDO_IN Current vs. LDO_IN Voltage, Internal LDO Enabled
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
6.5
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
6.0
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
5.5
5.0
4.5
4.0
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
3.5
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
3.0
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
ANALOG INPUT HIGH-Z MODE,
2.5
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
2.0
1.5
–40
–20
0
20
40
60
80
100
120
–40
–20
0
20
40
60
80
100
120
TEMPERATURE (°C)
TEMPERATURE (°C)
Figure 47. AVDD Current vs. Temperature
Figure 50. LDO_IN Current vs. Temperature, Internal LDO Enabled
Rev. 0 | Page 20 of 96
Data Sheet
AD4695/AD4696
5.0
300
250
200
150
100
50
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED
ANALOG INPUT HIGH-Z MODE,
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
REFERENCE INPUT HIGH-Z MODE ENABLED
0
0
100 200 300 400 500 600 700 800 900 1000
SAMPLE RATE (kSPS)
0
100 200 300 400 500 600 700 800 900 1000
SAMPLE RATE (kSPS)
Figure 51. VDD Current vs. Sample Rate, Internal LDO Disabled
Figure 54. VIO Current vs. Sample Rate, Conversion Mode,
OSR = 1
6.5
300
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
6.0
280
260
240
220
200
180
160
140
120
100
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
1MSPS
500kSPS
1.71 1.73 1.75 1.77 1.79 1.81 1.83 1.85 1.87 1.89
–40
–20
0
20
40
60
80
100
120
VDD VOLTAGE (V)
TEMPERATURE (°C)
Figure 52. VDD Current vs. VDD Voltage, Internal LDO Disabled
Figure 55. VIO Current vs. Temperature, Conversion Mode,
OSR = 1
6.5
120
ANALOG INPUT HIGH-Z MODE,
AVDD CURRENT
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
LDO_IN CURRENT (INTERNAL LDO ENABLED)
VDD CURRENT (INTERNAL LDO DISABLED)
VIO CURRENT
6.0
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
100
80
60
40
20
0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
–40
–20
0
20
40
60
80
100
120
–40
–20
0
20
40
60
80
100
120
TEMPERATURE (°C)
TEMPERATURE (°C)
Figure 53. VDD Current vs. Temperature, Internal LDO Disabled
Figure 56. Standby Current vs. Temperature
Rev. 0 | Page 21 of 96
AD4695/AD4696
Data Sheet
16
35
30
25
20
15
10
5
ANALOG INPUT HIGH-Z MODE,
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED
ANALOG INPUT HIGH-Z MODE,
14
12
10
8
REFERENCE INPUT HIGH-Z MODE ENABLED
REFERENCE INPUT HIGH-Z MODE ENABLED
6
4
2
0
0
0
100 200 300 400 500 600 700 800 900 1000
SAMPLE RATE (kSPS)
0
100 200 300 400 500 600 700 800 900 1000
SAMPLE RATE (kSPS)
Figure 57. Power Consumption vs. Sample Rate, Internal LDO Disabled
Figure 59. Power Consumption vs. Sample Rate, Internal LDO Enabled
18
16
14
40
35
30
ANALOG INPUT HIGH-Z MODE,
12
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 1MSPS
25
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
10
8
REFERENCE INPUT HIGH-Z MODE ENABLED, 500kSPS
20
6
15
4
10
ANALOG INPUT HIGH-Z MODE,
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
ANALOG INPUT HIGH-Z MODE,
2
REFERENCE INPUT HIGH-Z MODE DISABLED, 1MSPS
5
ANALOG INPUT HIGH-Z MODE,
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
REFERENCE INPUT HIGH-Z MODE DISABLED, 500kSPS
0
–40
0
–40
–20
0
20
40
60
80
100
120
–20
0
20
40
60
80
100
120
TEMPERATURE (°C)
TEMPERATURE (°C)
Figure 60. Power Consumption vs. Temperature, Internal LDO Enabled,
fS = 1 MSPS and 500 kSPS
Figure 58. Power Consumption vs. Temperature, Internal LDO Disabled,
fS = 1 MSPS and 500 kSPS
Rev. 0 | Page 22 of 96
Data Sheet
AD4695/AD4696
TERMINOLOGY
Integral Nonlinearity Error (INL)
Dynamic Range
INL is the maximum deviation from a straight line passing through
the endpoints of the ADC transfer function. The endpoints of the
transfer function are zero scale, a point ½ LSB below the first
code transition, and full scale, a point ½ LSB above the last code
transition.
Dynamic range is the ratio of the rms value of the full scale to
the total rms noise measured with the inputs shorted together.
The value for dynamic range is expressed in dB and is measured
with a signal at −60 dBFS to include all noise sources and DNL
artifacts.
Differential Nonlinearity Error (DNL)
Signal-to-Noise Ratio (SNR)
In an ideal ADC, code transitions are 1 LSB apart. DNL is the
maximum deviation from this ideal value. DNL is often specified
in terms of resolution for which no missing codes are guaranteed.
SNR is the ratio of the rms value of the actual input signal to
the rms sum of all other spectral components below the Nyquist
frequency, excluding harmonics and dc. The value for SNR is
expressed in dB.
Offset Error
Signal-to-Noise-and-Distortion (SINAD) Ratio
The offset error is the deviation of the measured transition
between −FSR and −FSR + 1 from the ideal transition, measured in
volts. The ideal transition between −FSR and −FSR + 1 occurs
at an analog input level ½ LSB above the IN− voltage (see the
Transfer Function section).
SINAD is the measured ratio of signal-to-noise-and-distortion
at the output of the ADC. The signal is the rms amplitude of
the fundamental. Noise is the sum of all nonfundamental
signals up to half the sampling frequency (fS/2), excluding dc.
Channel to -Channel Memory
Offset Error Match
Channel to channel memory is a measure of the level of crosstalk
that occurs when switching between channels in a channel
sequence. It is measured by applying a full-scale, 100 kHz signal
to one analog input channel and a dc voltage on another analog
input channel, and repeatedly switching between the two channels
between each conversion. The channel-to-channel memory is
the magnitude at 100 kHz in the spectrum measured from the
dc channel data.
Offset error match is the difference in offset error between any
two input channels.
Gain Error
The gain error is the deviation of the measured transition between
+FSR − 1 and +FSR from the ideal transition, and is measured in
percentage of full scale (%FS). The ideal transition between +FSR −
1 and +FSR occurs for an analog input level 1½ LSB below the
nominal full scale (see the Transfer Function section).
Channel to Channel Isolation
Gain Error Match
Gain error match is the difference in gain error between any
two input channels.
Channel to channel isolation is a measure of the level of crosstalk
from a signal on an inactive channel to an active channel. To
measure channel to channel isolation, apply a dc input to one
analog input channel and a full-scale, 100 kHz sine wave signal
to all other analog input channels and perform conversions
only on the dc input channel. The channel to channel isolation
is the magnitude at 100 kHz in the spectrum measured from
the dc channel data.
Spurious-Free Dynamic Range (SFDR)
SFDR is the difference, in decibels (dB), between the rms
amplitude of the input signal and the peak spurious signal.
Effective Number of Bits (ENOB)
ENOB is a measurement of the resolution with a sine wave
input and is related to SINAD by the following formula:
Total Harmonic Distortion (THD)
THD is the ratio of the rms sum of harmonics to the fundamental
and is defined as
ENOB = (SINAD − 1.76)/6.02
ENOB is expressed in bits.
2
2
2
2
2
V2 V3 V4 V5 V6
Noise Free Code Resolution
THD dB 20 log
V1
Noise free code resolution is the number of bits beyond which
it is impossible to distinctly resolve individual codes. To calculate
the resolution, use the following equation:
where:
V1 is the rms amplitude of the fundamental.
V2, V3, V4, V5, and V6 are the rms amplitudes of the second
Noise Free Code Resolution = log2 (2N/Peak-to Peak-Noise)
through the sixth harmonic.
Noise free code resolution is expressed in bits.
Aperture Delay
Aperture delay is the measure of the acquisition performance.
Aperture delay is the time between the rising edge of the CNV
input and when the input signal is held for a conversion.
Rev. 0 | Page 23 of 96
AD4695/AD4696
Data Sheet
THEORY OF OPERATION
contents. Conversion mode is used to initiate conversions and
read back conversion results. The fast conversion time of the
AD4695/AD4696 allows low serial clock rates to read back
conversions even when running at full throughput. The AD4695/
AD4696 support 4-wire SPI protocol and have optional dual-SDO
mode that enable slower SCK rates by shifting out conversion
results on multiple data outputs in parallel.
OVERVIEW
The AD4695/AD4696 are low power, 16-channel, 16-bit,
500 kSPS/1 MSPS, multiplexed, precision SAR ADCs. The
AD4695/AD4696 offer valid first conversion results even after
being idle for long periods of time.
The AD4695/AD4696 include features that simplify the design
requirements of peripheral circuitry and facilitate high
performance data acquisition system designs with low power
consumption and high channel density. These features include
the following:
The power consumption of the AD4695/AD4696 scales
with throughput because the ADC core powers down between
conversions. When operating at 10 kSPS, for example, the AD4695/
AD4696 typically consume 0.1 mW (with internal LDO, analog
input high-Z mode, and reference high-Z mode disabled),
making the devices suitable for battery-powered applications.
16-bit SAR ADC core with no missing codes
16 multiplexed analog inputs with low crosstalk multiplexer
Flexible channel sequencing modes
Analog input and reference high-Z mode
Temperature sensor
Input overvoltage protection clamps on each analog input
Programmable threshold detection for each analog input
Autocycle mode for performing conversions autonomously
First-order offset and gain correction for each analog input
Oversampling and decimation options for each analog input
The AD4695/AD4696 are available in a 32-lead, 5 mm × 5 mm
LFCSP.
CONVERTER OPERATION
The AD4695/AD4696 contain an SAR-based ADC core that
utilizes a charge redistribution digital-to-analog-converter
(DAC) to quantize the applied input voltage to an output code.
Figure 61 shows a simplified schematic of the AD4695/AD4696
SAR ADC core.
When multiplexing between channels, the analog input high-Z
mode feature reduces the nonlinear voltage steps that occur at
the analog inputs. Analog input high-Z mode relaxes settling
and bandwidth requirements of the analog front-end circuitry
and allows lower bandwidth and lower power amplifiers to
drive the analog inputs directly.
The analog inputs and the temperature sensor are connected to
the capacitor array inputs (ADCIN+ and ADCIN−) via the
internal low crosstalk multiplexer, represented by SWMUX+ and
SWMUX− in Figure 61. The multiplexer switches are controlled
by the internal channel sequencing logic and are updated once
per conversion (see the Multiplexer section and the Channel
Sequencing Modes section).
The reference input high-Z mode feature significantly reduces
the REF input current while the ADC core performs conversions
to relax the drive requirements of the reference circuitry. This
feature allows the use of lower power references and smaller
reference decoupling capacitors (1 μF) than with traditional
SAR ADCs.
The AD4695/AD4696 SAR ADC conversion routine consists of
an acquisition phase and a conversion phase. The ADC remains in
the acquisition phase until the conversion phase begins. During
the acquisition phase, the capacitor array acquires the voltage on
the analog input channel selected by the internal multiplexer.
During the conversion phase, the ADC core samples the input
voltage and generates a corresponding output code result.
Figure 62 shows the data processing path for the conversion
results generated by the AD4695/AD4696 ADC core.
Each analog input is equipped with input overvoltage
protection clamps to protect the device from overvoltage
events. The circuits of the clamps are robust and prevent
overvoltage events on an analog input from significantly
impacting the performance of the other analog inputs.
The AD4695/AD4696 must be in conversion mode to initiate
the conversion phase (see the Conversion Mode section). In
register configuration mode, the SAR ADC core remains in the
acquisition phase.
The AD4695/AD4696 include a variety of channel sequencing
modes that provide a flexible means of performing conversions
on a sequence of analog input channels. The standard sequencer
and advanced sequencer allow a channel sequence to be
preprogrammed and automatically progressed as conversions
occur. Two-cycle command mode and single-cycle command
mode allow the digital host to manually select from the
channels with SPI commands.
During the acquisition phase, the terminals of the capacitor
array tied to the input of the comparator are connected to
REFGND through the SW+ and SW− switches. All switches on
the individual capacitors in the array are connected to ADCIN+
and ADCIN−, and ADCIN+ and ADCIN− are connected to the
The AD4695/AD4696 have an enhanced digital interface that
is used to access the device register contents and initiate and read
conversion results while providing additional utility. Register
configuration mode is used to read and write to the register
selected analog input channel through SWMUX+ and SWMUX−
The acquisition phase ends immediately at the beginning of the
conversion phase.
.
Rev. 0 | Page 24 of 96
Data Sheet
AD4695/AD4696
The conversion phase is initiated by a rising edge on the CNV
input (in conversion mode only). When the conversion phase
begins, SW+, SW−, SWMUX+, and SWMUX− open first and sample
the analog input voltage on the capacitor arrays. The two capacitor
arrays are then disconnected from ADCIN+ and ADCIN− and
connected to REFGND. The sampled voltage is applied to the
comparator inputs, which causes the comparator to become
unbalanced. The ADC control logic performs a bit trial for each
capacitor in the array, starting with the MSB, by switching each
element of the capacitor array between REFGND and REF in
sequence. During each bit trial, the comparator input varies by
binary weighted voltage steps (VREF/2, VREF/4, …, VREF/65536),
and the control logic acts to bring the comparator back into a
balanced condition. The state of the comparator is recorded for
each bit trial to produce the resulting conversion result. The
conversion phase terminates when all bit trials are complete
and the conversion result is ready.
The delay between the end of each acquisition phase and the
beginning of the following acquisition phase depends on the
channel sequencing mode selected. When two-cycle command
mode, the standard sequencer, or the advanced sequencer are
enabled, the internal control logic determines the timing of the
start of the next acquisition phase. When single-cycle command
mode is enabled, the ADC core cannot enter the acquisition
phase until the 5-bit channel command is received over the SPI
(see the Single-Cycle Command Mode section).
The minimum acquisition time specification (tACQ) in Table 2
indicates the minimum amount of time that the AD4695/AD4696
are in the acquisition phase when running at the maximum
sample rate.
When analog input high-Z mode is disabled, the switches that
connect the analog inputs to the capacitor arrays close
immediately at the start of the acquisition phase. When analog
input high-Z mode is enabled, these switches close partway
through the acquisition phase, but the resulting voltage kickback is
significantly reduced. As a result, the settling time and bandwidth
requirements of the analog front-end circuitry are reduced
when analog input high-Z mode is enabled (see Figure 20 and
the Signal Settling Requirements section).
The SAR ADC core generates one output code for each conversion
phase. Multiple output codes are averaged together to generate
an oversampled ADC result when the active channel is configured
with an OSR setting greater than 1 (see the Transfer Function
section and Oversampling and Decimation section).
The conversion time specification (tCONVERT) in Table 2 refers to
the delay between a CNV rising edge and the end of the conversion
phase. During the conversion phase, the ADC generates a busy
indicator to communicate to the digital host when a conversion is
complete and ready to be read via the SPI (see the Busy Indicator
section). When enabled, the busy indicator transitions high at
the start of the conversion phase, and transitions low at the end
of the conversion phase.
The AD4695/AD4696 ADC core is controlled by an internal
clock, and the SPI serial clock (SCK) is not required for the
conversion process.
SW
MUX+
ADCIN+
IN+
SWITCHES CONTROL
SW+
LSB
MSB
32,768C 16,384C
4C
4C
2C
2C
C
C
C
BUSY
REF
REFGND
CONTROL
LOGIC
COMP
32,768C 16,384C
MSB
C
OUTPUT CODE
LSB
SW–
CNV
SW
MUX–
IN–
ADCIN–
Figure 61. ADC Simplified Schematic
Rev. 0 | Page 25 of 96
AD4695/AD4696
Data Sheet
for channels configured in pseudo bipolar mode (see the
Channel Configuration Options section).
TRANSFER FUNCTION
Figure 62 shows the AD4695/AD4696 data processing path. The
SAR ADC core generates one 16-bit output code per conversion
period. The OSR setting for the selected analog input channel
determines how many consecutive 16-bit output codes results
are averaged, and then the offset and gain correction settings are
applied to generate the final result to be read over the SPI in
conversion mode (see the Oversampling and Decimation section
and the Offset and Gain Correction section).
The AD4695/AD4696 include offset and gain correction for
each channel that can be configured to compensate for first-
order system errors. The offset and gain correction registers
modify the ADC transfer function digitally (see the Offset and
Gain Correction section).
The ideal transfer function is shown in Figure 63. The Converting
Between Codes and Volts section describes the relationship
between output codes and input voltages vs. VREF, OSR, polarity
modes, and offset and gain correction settings. Table 7 through
Table 10 show examples of different voltage inputs and the
corresponding results for each OSR and polarity mode option
(assuming an ideal ADC transfer function and with the offset
and gain correction values set to default values).
The conversion result length is determined by the OSR setting.
The conversion result resolution can range from 16 bits to 19 bits
for an OSR of 1 and 64, respectively (see the Oversampling and
Decimation section).
The conversion result encoding format is determined by the
selected polarity mode. The results are in straight binary format
for channels configured in unipolar mode, and twos complement
CNV
CONVERSION
RESULTS
OVERSAMPLING
ADC
OFFSET
AND
AND GAIN
CORE
SPI
DECIMATION
CORRECTION
BUSY
THRESHOLD
DETECTION
ALERT_STATUSn[7:0]
Figure 62. ADC Data Processing Path
TWOS
COMPLEMENT
STRAIGHT
BINARY
011...111 111...111
011...110 111...110
011...101 111...101
100...010 000...010
100...001 000...001
100...000 000...000
–FSR
–FSR + 1LSB
+FSR – 1LSB
+FSR – 1.5LSB
ANALOG INPUT
–FSR + 0.5LSB
Figure 63. ADC Ideal Transfer Function (FSR Is Full-Scale Range)
Table 7. Output Codes and Ideal Input Voltages, VREF = 5 V, OSR = 1
Input Voltage in
Unipolar Mode
Digital Output Code
(Straight Binary)
Input Voltage in Pseudo
Bipolar Mode
Digital Output Code (Twos
Complement)
Description
FSR − 1 LSB
Midscale + 1 LSB
Midscale
Midscale − 1 LSB
−FSR + 1 LSB
−FSR
4.999924 V
2.500076 V
2.5 V
2.499924 V
76.3 μV
0 V
0xFFFF
0x8001
0x8000
0x7FFF
0x0001
0x0000
2.499924 V
76.3 μV
0 V
−76.3 μV
−2.499924 V
−2.5 V
0x7FFF
0x0001
0x0000
0xFFFF
0x8001
0x8000
Rev. 0 | Page 26 of 96
Data Sheet
AD4695/AD4696
Table 8. Output Codes and Ideal Input Voltages, VREF = 5 V, OSR = 4
Input Voltage in
Unipolar Mode
Digital Output Code
(Straight Binary)
Input Voltage in Pseudo
Bipolar Mode
Digital Output Code (Twos
Complement)
Description
FSR − 1 LSB
4.999962 V
0x1FFFF
0x10001
0x10000
0x0FFFF
0x00001
0x00000
2.499962 V
38.1 μV
0 V
−38.1 μV
−2.499962 V
−2.5 V
0x0FFFF
0x00001
0x00000
0x1FFFF
0x10001
0x10000
Midscale + 1 LSB 2.500038 V
Midscale 2.5 V
Midscale − 1 LSB 2.499962 V
−FSR + 1 LSB
−FSR
38.1 μV
0 V
Table 9. Output Codes and Ideal Input Voltages, VREF = 5 V, OSR = 16
Input Voltage in
Unipolar Mode
Digital Output Code
(Straight Binary)
Input Voltage in Pseudo
Bipolar Mode
Digital Output Code (Twos
Complement)
Description
FSR − 1 LSB
4.999981 V
0x3FFFF3
0x20001
0x20000
0x1FFFF
0x00001
0x000004
2.499981 V
19.1 μV
0 V
−19.1 μV
−2.499981 V
−2.5 V
0x1FFFF
0x00001
0x00000
0x3FFFF
0x20001
0x20000
Midscale + 1 LSB 2.500019 V
Midscale 2.5 V
Midscale − 1 LSB 2.499981 V
−FSR + 1 LSB
−FSR
19.1 μV
0 V
Table 10. Output Codes and Ideal Input Voltages, VREF = 5 V, OSR = 64
Input Voltage in
Unipolar Mode
Digital Output Code
(Straight Binary)
Input Voltage in Pseudo
Bipolar Mode
Digital Output Code (Twos
Complement)
Description
FSR − 1 LSB
Midscale + 1 LSB
Midscale
Midscale − 1 LSB
−FSR + 1 LSB
−FSR
4.999910 V
2.500010 V
2.5 V
2.499990 V
9.54 μV
0 V
0x7FFFF3
0x40001
0x40000
0x3FFFF
0x00001
0x000004
2.499990 V
9.54 μV
0 V
−9.54 μV
−2.499990 V
−2.5 V
0x3FFFF
0x00001
0x00000
0x7FFFF
0x40001
0x40000
Each analog input has a unique overvoltage protection clamp
circuit, represented by OV clamp in Figure 64. The clamps
protect the analog inputs from dc overvoltage conditions and
eliminate the need for additional external protection diodes. See
the Input Overvoltage Protection Clamps section for a detailed
description of the overvoltage protection clamps.
ANALOG INPUTS
Figure 64 shows an equivalent circuit of the AD4695/AD4696
analog inputs (IN0 to IN15 and COM).
REF
D
1
IN0 TO IN15,
C
DAC
R
R
COM
IN
EXT
REXT and CEXT in Figure 64 represent an external, RC low-pass
filter, which is included in the system design to limit the bandwidth
of the input signal. REXT can also be used to improve overvoltage
protection of the analog inputs. See the External RC Filter section
for detailed descriptions of the REXT and CEXT functions.
C
C
PIN
OV
CLAMP
V
EXT
D
2
IN
REFGND
Figure 64. Equivalent Analog Input Circuit
A low crosstalk analog multiplexer routes the signals from the
analog input pins to the ADC core inputs. The impedance of the
analog inputs are modeled as the parallel combination of the
pin capacitance (CPIN) and the network formed by the series
connection of RIN and CDAC. RIN represents the ADC input series
resistance and the multiplexer switch resistance and is typically
240 Ω. CDAC represents the ADC sampling capacitive DAC shown
in Figure 61, and is typically 60 pF.
Rev. 0 | Page 27 of 96
AD4695/AD4696
Data Sheet
The two signal polarity modes are called unipolar mode and
pseudobipolar mode. When a channel is in unipolar mode, the
signal routed to ADCIN− is nominally 0 V (relative to REFGND).
When a channel is in pseudobipolar mode, the signal routed to
ADCIN− is nominally VREF/2 V (relative to REFGND). The
valid operating input voltage specification for unipolar and
pseudobipolar modes are shown in Table 1.
Multiplexer
The AD4695/AD4696 contain a flexible, low crosstalk analog
multiplexer for selecting from the 16 analog inputs and internal
temperature sensor and routing them to the inputs of the 16-bit,
pseudo differential SAR ADC core. Figure 65 shows a simplified
schematic of the internal multiplexer. The SWMUX+ and SWMUX−
switches shown in Figure 61 and Figure 65 represent the
multiplexer switches that route the selected channel to the ADC
inputs (labeled ADCIN+ and ADCIN− in Figure 61). SWMUX+
and SWMUX− are break-before-make and are controlled by the
internal channel sequencing logic (see the Channel Sequencing
Modes section).
When an input is configured in unipolar mode, its output codes
are in straight binary format. When an input is configured in
pseudobipolar mode, its output codes are in twos complement
format. See the Transfer Function section for an example of the
output code formatting for both unipolar and pseudobipolar
modes.
The multiplexer allows flexible analog input channel configuration.
The SWMUX− position is user programmable, and can be
assigned to any of the pins shown in Figure 65 (see the Channel
Configuration Options section).
The pin pairing assignments are selected with the IN_PAIR bit
field in the CONFIG_INn registers. The signal polarity modes are
selected with the IN_MODE field in the CONFIG_INn registers.
When an even numbered input is paired with its corresponding
odd numbered input, selecting the odd numbered input through
any of the channel sequencing modes is functionally identical
to selecting the even numbered input. The even numbered input is
always connected to ADCIN+, the odd numbered input is always
connected to ADCIN−, and only the settings in the even numbered
input CONFIG_INn register are applied. It is recommended to
only include the even numbered input in the channel sequence
when the input is assigned as part of a channel pair.
MUX+
IN0 TO IN15
TEMPERATURE
SENSOR
SAR
ADC
ODD
NUMBERED
INx
When the standard sequencer is enabled, the pin pairing
assignment settings are the same for all 16 analog inputs and are set
by the IN_PAIR field in the CONFIG_IN0 register. When the
advanced sequencer, two-cycle command mode, or single-cycle
command mode is enabled, the pin pairing assignment settings are
independent for all 16 analog inputs and are set by the IN_PAIR
field in the corresponding CONFIG_INn register for each input.
The polarity mode settings for each analog input are always set by
the IN_MODE bits in the corresponding CONFIG_INn registers,
regardless of the channel sequencing mode.
COM
REFGND
MUX–
Figure 65. Multiplexer Simplified Schematic
Channel Configuration Options
The AD4695/AD4696 feature several channel configuration
options that allow the device to interface with a variety of signals.
The channel configuration can be independently programmed
for each of the 16 analog inputs (IN0 through IN15).
Note that pseudobipolar mode is not available for channels
with the REFGND pin pairing assignment selected. If a channel
pin pairing assignment is configured as REFGND, the state of
the IN_PAIR field is ignored.
The channel configuration settings include pin pairing
assignments and signal polarity modes. The pin pairing options
assign the position of SWMUX− for each position of SWMUX+ and
determine which signal is routed to the negative side of the SAR
ADC core (ADCIN− in Figure 61). The signal polarity modes
configure the ADCIN− voltage range. Figure 66 shows the pin
pairing and voltage ranges for the different channel configuration
options.
INx
The pin pairing assignment options include the following:
INx+1
Figure 66, IN0 to IN15 paired with REFGND
Figure 67, IN0 to IN15 paired with COM
Figure 68, even numbered input paired with the next
highest odd numbered input (for example, IN0 with IN1,
IN2 with IN3, and so on).
COM
REFGND
Figure 66. Channel Configuration Option A
Rev. 0 | Page 28 of 96
Data Sheet
AD4695/AD4696
The reduction in the voltage kickback increases the effective
input impedance of the AD4695/AD4696 analog inputs and
reduces the bandwidth requirements of the AFE circuitry to
achieve desired settling accuracy and performance. The relaxed
bandwidth requirements of the AD4695/AD4696 simplify the
AFE circuit design by broadening the selection of compatible
amplifiers and external RC filter components. Therefore,
analog input high-Z mode helps remove the requirement of
dedicated ADC driver amplifiers per channel, which significantly
reduces system footprint and power consumption.
INx
INx+1
COM
0V
OR
V
/2
REF
REFGND
The analog input high-Z mode also reduces performance
degradation caused by series resistance between the front-end
amplifiers and the AD4695/AD4696 analog inputs, which allows
the resistor in the external RC filter (shown as REXT in Figure 64
and Figure 107) to be larger compared to traditional multiplexed
SAR ADCs. Using larger REXT with smaller CEXT alleviates amplifier
stability concerns without significantly impacting distortion
performance.
Figure 67. Channel Configuration Option B
INx
Figure 69 and Figure 70 demonstrate how a lower power, lower
bandwidth amplifier (ADA4077-1) can achieve the same ac
performance as a lower noise, higher bandwidth ADC driver
amplifier (ADA4807-1) by utilizing the AD4695/AD4696
analog input high-Z mode. Figure 69 and Figure 70 show the
SNR and THD performance of the AD4695/AD4696 paired
with the ADA4077-1 and ADA4807-1 with various external RC
filter components with analog input high-Z mode disabled and
enabled. Figure 71 shows the circuit configuration used to measure
the performance metrics shown in Figure 69 and Figure 70. The
standard sequencer is configured to alternate between two
AD4695/AD4696 channels once per conversion. The channels
are driven by antiphase, full-scale, 1 kHz sine waves.
0V
INx+1
OR
V
/2
REF
COM
REFGND
Figure 68. Channel Configuration Option C
Analog Input High-Z Mode
To achieve optimal data sheet performance from traditional
high resolution multiplexed SAR ADCs, system designers must
often include dedicated, high bandwidth, low noise ADC driver
amplifiers between the analog signal conditioning circuitry and
the ADC inputs to settle the voltage kickback that occurs at the
analog inputs between conversions. The AD4695/AD4696
analog input high-Z mode simplifies the design requirements
of the AFE circuitry that drives the analog inputs and facilitates
the design of small footprint, high channel density, precision
multiplexed SAR ADC signal chains.
The ADA4807-1 is a low noise, high bandwidth amplifier that
is typically recommended for driving precision SAR ADCs, and
the ADA4077-1 is a high precision, low drift amplifier with a
comparably lower bandwidth. Table 11 shows the bandwidth,
input noise, and supply current specifications for the ADA4807-1
and ADA4077-1. When analog input high-Z mode is disabled,
the ADA4077-1 THD performance is degraded because of its
inability to settle the voltage kickback between conversions.
When analog input high-Z mode is enabled, the ADA4077-1 is
able to achieve similar THD performance to the ADA4807-1,
despite having a comparably lower bandwidth. In the example
shown in Figure 71, analog input high-Z mode removes the
need for an ADA4807-1 or equivalent ADC driver amplifier for
each of its 16 analog input channels, which reduces the standby
current consumption of the system by roughly 16 mA, and
drastically reduces the full solution footprint.
Analog input high-Z mode significantly reduces the magnitude
of the voltage kickback that occurs at the analog inputs when
the ADC and multiplexer switches reconnect at the start of the
ADC acquisition phase (see the Signal Settling Requirements
section). Figure 20 shows the voltage kickback that occurs on
an analog input driven to 5 V after switching from another
analog input driven to 0 V with analog input high-Z mode
disabled and enabled.
Table 23 provides a list of recommended companion amplifiers
and external RC filter components to pair with the AD4695/
AD4696 for different target sample rates and input signal
bandwidths.
Rev. 0 | Page 29 of 96
AD4695/AD4696
Data Sheet
90
85
80
75
70
65
60
55
Analog input high-Z mode is enabled with the AINHIZ_EN bit
in the CONFIG_INn registers. When the standard sequencer is
enabled, analog input high-Z mode is enabled or disabled for all
16 analog inputs and is set by the AINHIZ_EN bit in the
CONFIG_IN0 register. When the advanced sequencer is enabled,
or when using two-cycle command mode or single-cycle
command mode, analog input high-Z mode is enabled or
disabled for all 16 analog inputs independently and is set by the
AINHIZ_EN bit in the corresponding CONFIG_INn register
for each input. Analog input high-Z mode is always enabled
when sampling the temperature sensor.
ADA4807-1, HIGH-Z MODE DISABLED, 500kSPS
ADA4807-1, HIGH-Z MODE DISABLED, 1MSPS
ADA4807-1, HIGH-Z MODE ENABLED, 500kSPS
ADA4807-1, HIGH-Z MODE ENABLED, 1MSPS
ADA4077-1, HIGH-Z MODE DISABLED, 500kSPS
ADA4077-1, HIGH-Z MODE DISABLED, 1MSPS
ADA4077-1, HIGH-Z MODE ENABLED, 500kSPS
ADA4077-1, HIGH-Z MODE ENABLED, 1MSPS
Analog input high-Z mode must be enabled when reference
input high-Z mode is enabled. If any analog input channels are
configured with analog input high-Z mode disabled, reference
input high-Z mode must also be disabled.
200Ω,
180pF,
4.42MHz
390Ω,
180pF,
2.27MHz
680Ω,
180pF,
1.30MHz
680Ω,
470pF,
498.0kHz
1.3kΩ,
470pF,
260.5kHz
EXTERNAL RC FILTER COMPONENTS, BANDWIDTH
Figure 69. SNR vs. External RC Filter Components, Bandwidth for Various
Amplifiers (VREF = 5 V, fIN = 1 kHz)
Table 11. Companion Amplifier Specifications
–35
ADA4807-1, HIGH-Z MODE DISABLED, 500kSPS
Input Voltage
Noise
−3dB
Supply Current
ADA4807-1, HIGH-Z MODE DISABLED, 1MSPS
Amplifier
Bandwidth per Amplifier
ADA4807-1, HIGH-Z MODE ENABLED, 500kSPS
–45
ADA4807-1, HIGH-Z MODE ENABLED, 1MSPS
ADA4807-1/ 3.1 nV/√Hz
ADA4807-2/
ADA4807-4
180 MHz
5.9 MHz
1.0 mA
400 μA
ADA4077-1, HIGH-Z MODE DISABLED, 500kSPS
ADA4077-1, HIGH-Z MODE DISABLED, 1MSPS
–55
ADA4077-1, HIGH-Z MODE ENABLED, 500kSPS
ADA4077-1, HIGH-Z MODE ENABLED, 1MSPS
–65
ADA4077-1/ 6.9 nV/√Hz
ADA4077-2/
ADA4077-4
–75
–85
–95
–105
–115
200Ω,
180pF,
4.42MHz
390Ω,
180pF,
2.27MHz
680Ω,
180pF,
1.30MHz
680Ω,
470pF,
498.0kHz
1.3kΩ,
470pF,
260.5kHz
EXTERNAL RC FILTER COMPONENTS, BANDWIDTH
Figure 70. THD vs. External RC Filter Components, Bandwidth for Various
Amplifiers (VREF = 5 V, fIN = 1 kHz)
AD4695/AD4696
CNV
R
EXT
IN0
SAR
ADC
C
EXT
ADA4807/ADA4077
ADA4807/ADA4077
INPUT
MUX
STANDARD
SEQUENCER
R
EXT
IN1
IN0
IN1
C
EXT
Figure 71. Amplifier and RC Filter Performance vs. Analog Input High-Z Mode Test Circuit
Rev. 0 | Page 30 of 96
Data Sheet
AD4695/AD4696
Overvoltage Protection Clamp Stability
INPUT OVERVOLTAGE PROTECTION CLAMPS
In applications where analog input overvoltage events are not a
concern, or in applications where clamp stability is not a
concern, the REXT and CEXT values are not required to follow the
guidelines described in this section.
The AD4695/AD4696 include overvoltage protection clamps on
IN0 to IN15 and COM to reduce the risk of device damage from
sustained dc overvoltage events. These clamps eliminate the
need for external clamping diodes in systems where the input
driving circuitry positive supply rail is greater than VREF (see
Figure 107).
The stability of the overvoltage protection clamp circuits depends
on the external RC filter component values and whether the
overvoltage reduced current mode is enabled or disabled. When a
clamp is unstable, it toggles between the active and inactive
states during overvoltage events. This instability causes small,
modulating currents to flow in both the overdriven input and the
reference, which can result in measurement errors in the
conversions of other analog inputs if the reference circuitry
does not have adequate load regulation to maintain a stable
reference voltage in response to the additional reference current.
Table 1 and Figure 42 show the additional reference input
(REF) current per active clamp.
Table 1 shows the activation, deactivation, and clamping voltages
of the overvoltage protection clamps. Figure 35 and Figure 38
show typical behavior of the clamps during overvoltage conditions.
The clamp circuits activate when the analog input voltage exceeds
the activation voltage. The clamps deactivate when the input
voltage drops below the deactivation voltage. While a clamp is
active, a flag is set in the status registers that can be read by the
digital host. See the Overvoltage Clamp Flags section for a detailed
description of the options for reading the status of each clamp.
The overvoltage protection clamps limit the extent to which input
overvoltage events disturb the reference source. When active, the
clamps limit the voltage on the analog inputs to the specified
clamping voltage and conduct the input current to ground
rather than through the ESD diode connecting the analog input
to the REF input (D1 in Figure 64), which prevents overvoltage
conditions on one analog input from degrading performance on
other analog inputs or other devices sharing the reference.
Figure 42 shows the relationship between a single clamp input
current and the resulting additional reference input current.
To ensure stable clamp operation, CEXT in the external RC filter
(as shown in Figure 64) must be at least 500 pF. The maximum
value of REXT is 1 kꢀ when the overvoltage reduced current
mode is enabled, and 2 kꢀ when the overvoltage reduced
current mode is disabled.
Overvoltage Clamp Flags
The AD4695/AD4696 provide several means to check the status
of the overvoltage protection clamps.
The INX_CLAMP_FLAG bits in the CLAMP_STATUS1 and
CLAMP_STATUS2 registers indicate the status of the overvoltage
protection clamps for IN0 to IN15. Each INX_CLAMP_FLAG bit
is asserted when the corresponding input clamp circuit is active
and is deasserted when the corresponding input clamp circuit is
inactive. The CLAMP_FLAG bit in the status register is asserted
when any combination of the overvoltage clamps on IN0 to IN15
are activated (when any of the INX_CLAMP_FLAG bits are
asserted). This bit is sticky and is only cleared when it is read while
all clamps are inactive.
Figure 26, Figure 29, and Figure 32 show the offset error, gain
error, and ac performance for one analog input channel vs. the
total number of active overvoltage protection clamps on the
other inputs.
Each overvoltage protection clamp circuit supports a maximum
sustained current of 5 mA. All 17 clamp circuits can sink 5 mA
simultaneously without damaging the device. The clamp current is
a function of VREF, the external series resistance (such as REXT in
Figure 64), and the output voltage of the AFE circuitry. See the
Input Overvoltage Protection Clamps section for details on
how to select REXT to prevent excess clamp current during
overvoltage events.
The COM_CLAMP_FLAG bit in the status register is asserted
when the COM input overvoltage protection clamp is active
and is deasserted when the COM input overvoltage protection
clamp is inactive. These bits can be read when in register
configuration mode to check the current status of each of the
overvoltage input clamp circuits.
Overvoltage Reduced Current Mode
The overvoltage reduced current mode further reduces additional
reference current during overvoltage events. Figure 42 shows
the difference between the additional reference input currents
drawn for different clamp input currents with the overvoltage
reduced current mode enabled and disabled.
The OV_ALT flag in the optional status bits allows all
overvoltage clamp statuses to be checked while performing
conversions. The OV_ALT flag is the bitwise logical OR of the
16 INX_CLAMP_FLAG bits in the CLAMP_STATUS1 and
CLAMP_STATUS2 registers. The OV_ALT flag can also be
configured as the logical OR of the overvoltage clamp flags and
the general threshold alert indicator (as described in the
Threshold Detection and Alert Indicators section). See the
Status Bits section for details on configuring the OV_ALT flag.
The overvoltage reduced current mode is enabled when the
OV_MODE bit in the REF_CTRL register is set to 0.
Overvoltage reduced current mode is enabled by default.
Enabling overvoltage reduced current mode changes the
maximum value of REXT and achieves stable clamp operation.
See the Overvoltage Protection Clamp Stability section for
more information on the relationship between the external RC
filter and clamp operation.
Rev. 0 | Page 31 of 96
AD4695/AD4696
Data Sheet
and drive capabilities, or the use of a dedicated reference buffer
to drive the REF input with a large reference decoupling capacitor.
See the Reference Circuitry Design section for more information
on properly selecting reference circuitry components.
TEMPERATURE SENSOR
The AD4695/AD4696 include a temperature sensor that
converts the die temperature to an output voltage that can be
sampled and converted to an output code by the SAR ADC core.
The relationship between the measured die temperature (T) and
the temperature sensor output voltage (VTEMP) is nominally
The AD4695/AD4696 incorporate features that simplify design
of the companion reference circuitry, and facilitate the design
of small footprint, low power systems. The reference input
high-Z mode reduces the REF input current by approximately
95%, allowing a broader selection of voltage references and
amplifiers to drive the REF input without impacting performance
(see the Reference Input High-Z Mode section).
mV
VTEMP 1.8
T 725 mV
C
The temperature sensor sensitivity is a measure of the change
in output voltage in relation to a change in device temperature,
and is typically −1.8 mV/°C. At 0°C, the temperature sensor
output is typically 725 mV.
The reference input current scales with sample rate (see Table 1
and Figure 39).
When the temperature sensor is selected, the multiplexer
SWMUX+ switch (see Figure 65) selects the temperature sensor
output and its SWMUX− switch selects REFGND, and the SAR
ADC core samples VTEMP to generate a corresponding output
code. The analog-to-digital conversion of the temperature
sensor output utilizes the same transfer function as an analog
input configured in unipolar mode with OSR = 1 (see the
Transfer Function section).
Reference Input High-Z Mode
When enabled, reference input high-Z mode reduces the average
REF current by approximately 95% from 320 μA/MSPS to 11
μA/MSPS. The reduction in REF current allows the AD4695/
AD4696 to tolerate larger series resistance between the reference
source and the REF input without compromising performance.
Therefore, reference input high-Z mode allows voltage references
with higher load regulation specifications to directly drive the REF
input without the need for a dedicated reference buffer.
When the standard sequencer or advanced sequencer is
enabled, the temperature sensor is sampled at the end of the
preprogrammed channel sequence if the TEMP_EN bit in the
TEMP_CTRL register is set to 1.
The REF input requires a reference decoupling capacitor (CREF).
When reference input high-Z mode is disabled, CREF must be
10 μF or larger. When reference input high-Z mode is enabled,
When using either two-cycle command mode or single-cycle
command mode, the temperature sensor can be selected by
writing the code 0x0F on SDI on the first five rising edges of
SCK in the same way analog inputs are selected (see Table 16).
C
REF can be as small as 1 μF.
See the Reference Circuitry Design section for more reference
circuit design recommendations.
To enable and disable reference input high-Z mode, set the
value of the REFHIZ_EN bit in the REF_CTRL register.
Reference input high-Z mode is enabled by default.
When the temperature sensor is enabled, analog input high-Z
mode is always enabled and the OSR is always 1. The
temperature sensor does not have threshold detection alerts.
Analog input high-Z mode must be enabled when reference
input high-Z mode is enabled. If any analog input channels are
configured with analog input high-Z mode disabled, reference
input high-Z mode must also be disabled.
VOLTAGE REFERENCE INPUT
VREF sets the ADC full-scale voltage (see the Transfer Function
section). The ADC core samples the voltage on the reference input
(REF) during the bit trials in the conversion process to determine
the output code result. The AD4695/AD4696 are compatible with
reference voltages from 2.4 V to 5.1 V.
POWER SUPPLIES
The AD4695/AD4696 have three power supply pins: an analog
supply (AVDD), an ADC core supply (VDD), and a digital
input/output interface supply (VIO). The AD4695/AD4696
also include an internal LDO that can be used to provide the
VDD rail with a wider variety of supply voltages (or in single-
supply systems by tying LDO_IN to AVDD). Table 1 shows the
specified power supply voltage requirements.
The AD4695/AD4696 must be configured for optimal
performance with the selected reference voltage. The VREF_SET
field in the REF_CTRL register provides five VREF range options,
as shown in Table 46. This value must be programmed to
match the VREF voltage applied to the REF pin.
A common challenge presented by traditional SAR ADCs is in
designing reference circuitry with sufficient drive capability to
maintain a precise VREF while the REF input dynamically draws
input current during the SAR bit trials. Deviations in VREF result
in reduction in ADC accuracy and performance, such as higher
gain error or distortion. The REF input presents a dynamic load as
the input pulls charge from the external reference circuitry at
different times in the SAR process. This process traditionally
requires either voltage references with sufficient load regulation
AVDD can range from 3.15 V to 5.5 V and powers the analog
front-end features of the AD4695/AD4696, including the analog
input high-Z mode and reference input high-Z mode circuitry.
VDD is nominally 1.8 V, and powers both the ADC core and the
device register memory. When power is first applied to VDD, the
ADC core initializes and the device register contents are set to the
default states (as shown in the Register Information section).
Rev. 0 | Page 32 of 96
Data Sheet
AD4695/AD4696
VIO can range from 1.2 V to 1.8 V and sets the input and
output levels for the digital interface pins. VIO allows direct
interfacing with digital controller logic levels between 1.2 V and
1.8 V (see the Digital Interface section for more information).
and is identical to performing a software reset (see the Device
Reset section for detailed descriptions of hardware and software
resets). The digital interface requires that VIO still be supplied
to accept the wake-up command, and the internal LDO will not be
enabled if VIO is not within the specified range (see Table 1).
Decouple AVDD to AGND and VIO to IOGND with at least
100 nF and decouple VDD to AGND with at least 1 μF. When
not using the internal LDO to supply VDD, LDO_IN does not
require decoupling.
OVERSAMPLING AND DECIMATION
The AD4695/AD4696 include an oversampling and decimation
engine that averages consecutive ADC samples to generate an
oversampled result with higher effective resolution and lower
effective noise (see Table 1).
The AD4695/AD4696 are independent of the power supply
sequencing between VIO, VDD, and AVDD (and LDO_IN when
the internal LDO is enabled). When VIO and VDD are first
supplied, a power-on reset (POR) initiates (see the Device Reset
section). Additionally, the AD4695/AD4696 are insensitive to
power supply ripple over a wide frequency range, as shown in
Figure 34.
Each analog input channel can be configured with an OSR of 1, 4,
16, or 64. Conversion results generated for channels with an OSR
of 4, 16, or 64 are 17 bits, 18 bits, or 19 bits long, as shown in
Table 19 and Table 20 and the Transfer Function section.
When a given analog input channel is selected by the channel
sequencing logic, the multiplexer continues to select that channel
until the specified number of conversions have been performed,
and the results of each of those conversions are averaged together
to generate a single output code. For example, if IN0 is configured
with an OSR of 64, one averaged result is produced after the 64th
consecutive CNV rising edge (when the AD4695/AD4696 are
in conversion mode). Configuring a channel with an OSR of 1
is equivalent to performing no oversampling on that channel.
Internal LDO
To minimize the number of system supply rails required to
power the AD4695/AD4696, the internal LDO can be used to
supply the VDD voltage internally. LDO_IN can be tied to
AVDD to enable a single supply to power the entire device
(excluding VIO, which must be powered by the digital host
input/output voltage).
To enable the internal LDO, LDO_IN must be driven to at least
2.4 V and VIO must already be powered. To enable the internal
LDO, set the LDO_EN bit in the setup register to 1. The internal
LDO is enabled by default on device power-up and after device
resets. When the internal LDO is enabled, its output drives
VDD internally. When the internal LDO is disabled, its output
is high impedance.
When enabled on the BSY_ALT_GP0 pin or the serial data
output(s), the busy indicator acts as a data ready signal, and
only transitions low when the oversampled result is available
(see the Busy Indicator section). Figure 75 shows the relative
timing of the busy indicator when the OSR for a channel is set
to a value other than 1.
It is not possible to power the VIO supply with the internal
LDO output. VIO must be supplied by the digital host or other
system supply rail.
The effective sample period of a given channel is equal to the
conversion period (tCYC) in Table 2 multiplied by its OSR. Figure 75
shows the relative timing of the CNV signal and the availability
of the oversampled result. Consider the OSR of each channel
when designing the channel sequence to achieve the desired
certain effective sample rates for each channel (see the Effective
Channel Sample Rate section).
When using the internal LDO, VDD must be floating, and the
VDD supply voltage is driven by the internal LDO output
automatically when LDO_IN and VIO are supplied. When not
using the internal LDO, LDO_IN must be tied to AGND and
VDD must be supplied externally.
The OSR is configured via the OSR_SET fields in the
CONFIG_INn registers (see Table 54).
The internal LDO output is designed to withstand being
powered up with VDD either driven by a separate 1.8 V supply
or inadvertently shorted to AGND. It is recommended to
ensure VDD is disconnected from any other rails or loads. The
internal LDO is not intended to power additional devices. It is
recommended to clear the LDO_EN bit when powering VDD
externally, even if the LDO_IN input is shorted to AGND (see
the Device Configuration Recommendations section).
When the standard sequencer is enabled, the OSR for all analog
input channels is the same and is set by the OSR_SET field in
the CONFIG_IN0 register. When the advanced sequencer is
enabled, each of the 16 analog input channels can be configured
with different OSR settings with the OSR_SET fields in the
corresponding CONFIG_INn registers.
Oversampling is not supported in two-cycle command mode or
single-cycle command mode. Set the OSR_SET fields for all
active channels to 0x0 when using two-cycle command mode
or single-cycle command mode.
The internal LDO can be disabled to put the AD4695/AD4696
in a low power state without disabling the AVDD, LDO_IN, or
VIO rails. When the internal LDO is disabled while VDD is not
powered by an external supply, the ADC core shuts down and
configuration register contents are erased. The internal LDO
can be enabled again either with a wake-up command over the
SPI, or with a hardware reset. The wake-up command is 0x81
Rev. 0 | Page 33 of 96
AD4695/AD4696
Data Sheet
When autocycle mode is enabled, the conversion signal is
generated internally by the AD4695/AD4696, and the
oversampling engine continues to wait for OSR conversion
periods before generating an output result.
THRESHOLD DETECTION AND ALERT INDICATORS
The AD4695/AD4696 include a threshold detection feature
with alert indicators that notify the digital host system when a
conversion result violates user defined upper and lower limits.
OFFSET AND GAIN CORRECTION
The TD_EN bit in the CONFIG_INn registers enables or disables
threshold detection for the corresponding analog input. When the
standard sequencer is enabled, threshold detection is enabled or
disabled for all analog inputs with the TD_EN bit in the
CONFIG_IN0 register. When the advanced sequencer, two-cycle
command mode, or single-cycle command mode is enabled,
threshold detection is enabled or disabled for each analog input
independently with the TD_EN bit in each of the corresponding
CONFIG_INn registers.
The AD4695/AD4696 include offset and gain error correction
functionality to correct for first-order nonidealities in a full
analog front-end signal chain. Offset and gain error correction
digitally adjusts the offset and gain of the overall ADC transfer
function (see the Transfer Function section).
The final output code is calculated with the following expression:
OUT = (IN + B) × M
where:
When threshold detection is enabled for a given analog input, the
ADC results generated for that analog input are compared against
an upper threshold value and lower threshold value. Upper and
lower threshold values can be independently assigned for each
of the 16 analog inputs. The upper and lower threshold values
for the 16 analog inputs are set with the upper and lower fields in
the UPPER_INn and LOWER_INn registers. The upper and
lower fields are 12 bits wide and correspond to the 12 MSBs of
the ADC results for all OSR options. For example, setting the
upper field to 0xFFF corresponds to an upper threshold value of
0xFFF0 when the OSR of that channel is 1, and 0x7FF80 when the
OSR of that channel is 64 (see the Oversampling and Decimation
section).
OUT is the final output code result.
IN is the result generated by the ADC (after oversampling).
B is the offset correction value.
M is the gain correction value.
The gain correction value (M) for each analog input is set with
the gain field in the corresponding GAIN_INn register. The
gain field is 16 bits wide and is in straight binary format. The
range of gain correction values is 0 to 1.99997, and is calculated
with the following expression:
M = Gain/215
where Gain is the value written to the gain field.
The offset correction value (B) for each analog input is set with
the offset field in the corresponding OFFSET_INn register. The
offset field is 16 bits wide and is in twos complement format to
enable positive and negative offset correction. The range of
offset correction values is FSR/8 for all OSR options, which
means the MSB of the offset field always corresponds to the MSB
– 3 bit of the ADC result. For example, when the OSR for a
given analog input channel is 1, the offset correction value is
equal to offset[15:3], and when the OSR is 64, the offset
correction value is offset[15:0]. Table 12 shows the offset
correction value for each OSR option.
When an analog input is configured in unipolar mode, the
corresponding upper and lower fields are in straight binary format.
When an analog input is configured in pseudobipolar mode,
the corresponding upper and lower fields are in twos complement
format.
Alert Indicator Registers
The ALERT_STATUS1 to ALERT_STATUS4 registers contain
the upper alert indicators (HI_INn) and lower alert indicators
(LO_INn) for all 16 analog inputs. The TD_ALERT bit in the
status register is the logical OR of the HI_INn and LO_INn bits.
When the ADC result is greater than or equal to the upper
threshold value, the corresponding HI_INn flag is set to 1.
When the ADC result is less than or equal to the lower threshold
value, the corresponding LO_INn flag is set to 1. When the OSR
of an INn analog input is greater than 1, the state of its
corresponding HI_INn and LO_INn flags update after the
oversampled result is generated.
Offset and gain correction are always enabled for all analog
input channels. When the OFFSET field for a given analog
input is set to 0x0000, the offset correction value is 0 and is
equivalent to applying no offset correction. When the GAIN
field for a given analog input is set to 0x8000, the gain correction
value is 1 and is equivalent to applying no gain correction.
Table 12. Oversample Ratio vs. Offset Correction Value
Reading the TD_ALERT bit indicates to the digital host
whether any upper or lower threshold was violated and reading
the HI_INn and LO_INn bits indicates which specific type of
threshold was violated on which channel. The AD4695/AD4696
must be in register configuration mode to read from the
registers that contain these alert indicator bits, but the state of
TD_ALERT can also be read via the status bits or the
Oversample Ratio
Offset Correction Value (B)
1
Offset[15:3]
4
Offset[15:2]
16
64
Offset[15:1]
Offset[15:0]
BSY_ALT_GP0 pin when these options are enabled (see the
Status Bits section and Alert Indicator on BSY_ALT_GP0 section).
Rev. 0 | Page 34 of 96
Data Sheet
AD4695/AD4696
The HI_INn and LO_INn bits are read to clear bits and are
automatically reset to 0 after being read in a SPI read
transaction (in register configuration mode).
Alert Indicator on BSY_ALT_GP0
When the alert indicator is enabled on BSY_ALT_GP0, the
state of the TD_ALERT bit is driven on the BSY_ALT_GP0 pin,
which allows threshold violations to be detected without
interrupting conversions. The combination of the alert
indicator on the BSY_ALT_GP0 pin and autocycle mode allows
the digital host serial interface to remain idle until a threshold
violation is detected (see the Autocycle Mode section).
When the ALERT_MODE bit in the setup register is set to 0,
the HI_INn and LO_INn bits also automatically clear based on
user programmable hysteresis settings. The hysteresis fields in
the 16 HYST_INn registers set the hysteresis value for the
corresponding analog input. Each analog input can be
programmed with different hysteresis values. When this option
is selected, each HI_INn bit automatically clears when the
corresponding analog input generates a conversion result that is
less than the upper threshold value minus the hysteresis value.
Each LO_INn bit automatically clears when the corresponding
analog input generates a conversion result that is greater than
the lower threshold value plus the hysteresis value. Figure 72
shows how the HI_INn and LO_INn bits are set and cleared
when ALERT_MODE is set to 0 and 1 as conversion results are
generated on the corresponding analog input channel.
ALERT_MODE is set to 0 by default.
Figure 91 through Figure 98 show the relative timing of CNV
rising edges and when the state of the BSY_ALT_GP0 pin is
configured as the alert indicator is updated.
Set the ALERT_GP_EN bit in the GP_MODE register to 1 to
enable the alert indicator on the BSY_ALT_GP0 pin.
The BSY_ALT_GP0 pin can also be configured to perform
other functions than the alert indicator, and all other higher
priority functions must be disabled to configure them as the
busy indicator. See the General-Purpose Pin section for details
on configuring the BSY_ALT_GP0 pin functionality.
OUTPUT CODES
0xC000
0xA000
UPPER = 0xC00
HYSTERESIS = 0x200
LOWER = 0x400
SAMPLES
0x6000
0x4000
HI_INn
ALERT_MODE = 1
LO_INn
TD_ALERT
HI_INn
LO_INn
ALERT_MODE = 0
TD_ALERT
Figure 72. Alert Indicator Behavior with Hysteresis Enabled and Disabled (Unipolar Mode, OSR = 1)
Rev. 0 | Page 35 of 96
AD4695/AD4696
Data Sheet
BUSY INDICATOR
CHANNEL SEQUENCING MODES
The busy indicator acts as a data ready signal that can be used
to trigger an interrupt service routine on the digital host to
initiate an SPI transaction to read the ADC result (see the
Conversion Mode section and SPI Peripheral Synchronization in
Conversion Mode section). The busy indicator can be enabled on
the serial data outputs and on the BSY_ALT_GP0 pin.
In conversion mode, the AD4695/AD4696 multiplexer channel
updates once per conversion period at the start of the ADC
core acquisition phase, as described in the Converter Operation
section. The multiplexer is controlled by internal channel
sequencing logic, and there are four options for programming
the channel sequence.
Busy Indicator on Serial Data Outputs
The standard sequencer and advanced sequencer automates
progression through a preprogrammed channel sequence.
When either the standard sequencer or advanced sequencer is
enabled, the digital host is not required to provide channel
sequencing instructions while reading conversion results over
the SPI, which reduces the digital resource requirements.
When the busy indicator is enabled on the serial data outputs,
the serial data outputs are high impedance while the ADC is in
the conversion phase, and transition low when the ADC result
is ready. Set the SDO_STATE bit in the setup register to 1 to
enable the busy indicator on the serial data outputs.
Two-cycle command mode and single-cycle command mode allow
the digital host to directly control the channel sequence via 5-bit
commands written over the serial interface during conversion
data readback frames. Two-cycle command mode and single-
cycle command mode enable systems with dynamic and
adaptive channel sequencing requirements, such as control
loop applications.
Figure 91 through Figure 98 show the relative timing of CNV
rising edges to the busy indicator on the serial data output(s).
The serial data output mode selected by the SDO_MODE field
determines which pins are assigned as serial data outputs (see the
Serial Data Output Modes section). When SDO_STATE is set
to 1, the busy indicator is enabled on all pins assigned as serial data
outputs. When single-SDO mode is selected, the busy indicator is
only output on SDO. When dual-SDO mode is selected, the
busy indicator is output on both SDO and BSY_ALT_GP0.
Figure 73 through Figure 77 show conversion mode example
timing diagrams of the AD4695/AD4696 multiplexer channel
selection, ADC sampling, and conversion data output relative
to the channel sequencing settings and the CNV signal. The
signal labeled busy refers to the busy indicator, which can be
enabled on either the BSY_ALT_GP0 pin or the serial data
output(s), as described in the Busy Indicator section. The signal
labeled SDOx refers to the SDO pin plus the BSY_ALT_GP0
pin if dual-SDO mode is enabled, as described in the Serial
Data Output Modes section.
When enabling the busy indicator on the serial data outputs,
place pull-up resistors (2 kꢀ minimum) on each utilized pin to
ensure that the serial data output lines are pulled high until the
ADC result is ready.
The serial data outputs are forced to a high impedance state
CS
CS
whenever the
pin is driven high. If the
pin is high when
the ADC result is ready, the serial data outputs remain high
CS
impedance until the
Interface section).
pin is brought low (see the Digital
Table 13 shows the configuration settings used to select from
the four channel sequencing modes. Both the STD_SEQ_EN bit
and the NUM_SLOTS_AS field are located in the SEQ_CTRL
register. The CYC_CTRL bit is located in the setup register.
Busy Indicator on BSY_ALT_GP0
When the busy indicator is enabled on the BSY_ALT_GP0 pin,
BSY_ALT_GP0 is driven high while the ADC is in the conversion
phase, and transitions low when the ADC result is ready. Set
the BUSY_GP_EN bit in the GP_MODE register to 1 to enable
the busy indicator on BSY_ALT_GP0.
As noted in the Channel Configuration Options section, when
even and odd numbered inputs are paired, selecting the odd
numbered input using any of the four channel sequencing
modes results in the same behavior as if the even numbered
input were selected instead. For this reason, it is recommended
to only include the even numbered input in the channel sequence.
Figure 91 through Figure 98 show the relative timing of CNV
rising edges to the busy indicator rising and falling edges.
Standard Sequencer
When the BSY_ALT_GP0 pin is assigned as the busy indicator,
the BSY_ALT_GP0 pin is not forced to high impedance when
The standard sequencer automates progression through a
preprogrammed set of enabled channels. The standard
sequencer is the simplest of the four channel sequencing modes
and is ideal for systems with fixed, static channel sequences.
CS
the
pin is high, which allows the digital host to leave the
serial interface completely disabled until a busy indicator falling
edge is registered (see the SPI Peripheral Synchronization in
Conversion Mode section).
The standard sequencer advances through each enabled
channel in ascending order and repeats the sequence until the
device exits conversion mode. The multiplexer channel is
updated to the next enabled channel each time a conversion
result is ready. Figure 73 shows an example where the standard
sequencer, three analog inputs (IN0, IN2, and IN15), and the
The BSY_ALT_GP0 pin can also be configured to perform other
functions than the busy indicator, and all other higher-priority
functions must be disabled to configure the BSY_ALT_GP0 pin as
the busy indicator. See the General-Purpose Pin section for
details on configuring the BSY_ALT_GP0 pin.
Rev. 0 | Page 36 of 96
Data Sheet
AD4695/AD4696
temperature sensor are enabled in the sequence with no
oversampling on any channel.
Advanced Sequencer
The advanced sequencer automates progression through a
preprogrammed channel sequence where the order of channels is
completely customizable. The advanced sequencer enables highly
flexible sequences of channels with minimal digital overhead.
The bits in the STD_SEQ_CONFIG register control which
channels are included in the channel sequence when the standard
sequencer is enabled. Each bit in the STD_SEQ_CONFIG register
corresponds to one of the 16 analog inputs, and each channel is
enabled if its corresponding bit is set to 1. If the TEMP_EN bit
in the TEMP_CTRL register is set to 1, the temperature sensor
is added to the end of the sequence as well. For the example in
Figure 73, the value programmed into the STD_SEQ_CONFIG
register is 0x1005, and the TEMP_EN bit is set to 1.
The advanced sequencer steps through a set of channel slots,
where each slot can be assigned to any of the 16 analog inputs
and sequences can be between two and 128 slots. The sequence
progresses through the enabled slots in ascending order starting
from Slot 0, and the sequence is repeated until the device exits
conversion mode. Figure 74 shows an example where the
advanced sequencer is enabled with four slots enabled and
assigned to IN6, IN10, IN6, and IN3 with the temperature
sensor enabled (with no oversampling on any channel).
To enable the standard sequencer, set the STD_SEQ_EN bit in
the SEQ_CTRL register to 1 and set the CYC_CTRL bit in the
setup register to 0 (see Table 13). The standard sequencer is
enabled by default.
The number of slots in the sequence is set with the NUM_
SLOTS_AS field in the SEQ_CTRL register. Each slot channel
assignment is set with the SLOT_INX fields in the AS_SLOTn
registers (located at Register Address 0x100 to Register
Address 0x17F), where AS_SLOT0 corresponds to Slot 0,
AS_SLOT1 corresponds to Slot 1 and so on. Table 60 shows the
values of SLOT_INX for each of the 16 analog inputs.
While the AD4695/AD4696 are in register configuration mode
when the STD_SEQ_EN bit in the SEQ_CTRL register is set to
1, the multiplexer automatically connects the first enabled
channel in the sequence to the ADC core inputs, which allows
the ADC to acquire the signal on that channel even before the
device enters conversion mode.
If the TEMP_EN bit in the TEMP_CTRL register is set to 1, the
temperature sensor is appended to the end of the sequence. The
temperature sensor cannot be selected with the SLOT_INX
fields in the AS_SLOTn registers.
When the standard sequencer is enabled, the control bits in the
CONFIG_IN0 register determine the configuration settings for
all INn analog inputs (except for the polarity mode, which is set
for each INn analog input independently with the IN_MODE
bit in the corresponding CONFIG_INn register). Therefore, all
analog inputs have the same pin pairing options, analog input
high-Z mode enable settings, OSR settings, and threshold
detection enable settings.
To enable the advanced sequencer, set the STD_SEQ_EN bit to
0, set the CYC_CTRL bit to 0, and set the NUM_SLOTS_AS
field to any value between 1 and 127 (see Table 13).
While the AD4695/AD4696 are in register configuration mode
when the STD_SEQ_EN bit in the SEQ_CTRL register is set to
0, the multiplexer automatically connects the channel specified
in the AS_SLOT0 register to the ADC core inputs, which allows
the ADC to acquire the signal on that channel even before the
device enters conversion mode.
The multiplexer does not advance to the next channel in the
sequence until the required number of conversions dictated by
the selected channel OSR setting is complete. For example, if
the OSR is set to 16, 16 CNV rising edges are required before
the conversion result is ready and the multiplexer selects the
next channel in the sequence. Figure 74 shows an example
timing diagram where the OSR for all channels is set to N. See
the Oversampling and Decimation section for more
information.
When the advanced sequencer is enabled, the configuration
settings for each channel are set with the corresponding
CONFIG_INn register. Therefore, all analog inputs can have
different channel configuration options, analog input high-Z
mode enable settings, OSR settings, and threshold detection enable
settings. Configure each CONFIG_INn register before entering
conversion mode and initiating conversions.
When the standard sequencer is enabled, each enabled analog
input is sampled once per sequence iteration, which means
each analog input has the same effective sample rate. See the
Effective Channel Sample Rate section for more information.
The multiplexer does not advance to the next channel in the
sequence until the required number of conversions dictated by
the selected channel OSR setting is complete. When the OSR of
a channel in the sequence is set to a value other than 1 (when
the OSR_SET field in the corresponding CONFIG_INn register is
not set to 0x0), the advanced sequencer does not advance to the
next channel in the sequence and the busy indicator does not
transition low until the required number of conversions is
complete. For example, if the OSR is set to 16, 16 CNV rising
edges are required before the conversion result is ready and the
multiplexer selects the next channel in the sequence. Figure 75
Table 13. Register Settings for Channel Sequencing Modes
Channel Sequencing
Mode
STD_SEQ_EN NUM_SLOTS_AS CYC_CTRL
Two-Cycle Command
Mode
0
0x00
0
Single-Cycle
0
0x00
1
Command Mode
Standard Sequencer
Advanced Sequencer
1
0
Don’t care
0
0
0x01 to 0x7F
Rev. 0 | Page 37 of 96
AD4695/AD4696
Data Sheet
shows an example timing diagram where OSR for IN0 is set to
N. See the Oversampling and Decimation section for more
information.
When two-cycle command mode is enabled, the configuration
settings for each channel are set with the corresponding
CONFIG_INn register. Therefore, all analog inputs can have
different channel configuration options, analog input high-Z
mode enable settings, and threshold detection enable settings.
Configure each CONFIG_INn register before entering
conversion mode and initiating conversions.
When the advanced sequencer is enabled, the channel sequence
can be configured to achieve different effective sample rates for
each channel. See the Effective Channel Sample Rate section for
more information.
Oversampling is not supported when two-cycle command
mode is enabled. Set the OSR for all analog inputs to 1 before
entering conversion mode with two-cycle command mode
enabled (see the Oversampling and Decimation section).
Two-Cycle Command Mode
Two-cycle command mode allows the digital host system to
manually control the next channel in the sequence on-the-fly
and enables dynamic channel sequencing without interrupting
conversions.
Single-Cycle Command Mode
Single-cycle command mode allows the digital host system to
manually control the next channel in the sequence on-the-fly
and enables dynamic channel sequencing without interrupting
conversions.
In two-cycle command mode, the channel sequence is determined
by 5-bit commands transmitted from the digital host during
conversion result readback frames. The 5-bit commands are
clocked in on SDI on the first five SCK rising edges in the frame
and latched into memory on the sixth SCK falling edge in the
frame. If a valid channel command is received, the conversion
result for that channel is available after two conversion periods.
Figure 76 shows the relative timing between the 5-bit
commands (represented by CMD) and the corresponding
acquisition phase, conversion phase, and conversion result
readback in two-cycle command mode.
In single-cycle command mode, the channel sequence is
determined by 5-bit commands transmitted from the digital
host during conversion result readback frames. The 5-bit
commands are clocked in on SDI on the first five SCK rising
edges in the frame and latched into memory on the sixth SCK
falling edge in the frame.
If a valid channel command is received, the conversion result
for that channel is available in only one conversion period.
Figure 77 shows the relative timing between the 5-bit
commands (represented by CMD) and the corresponding
acquisition phase, conversion phase, and conversion result
readback in single-cycle command mode.
Two-cycle command mode maximizes the acquisition time for
all channels because the 5-bit channel commands are latched in
before the multiplexer switches select the corresponding
channel and begin the ADC acquisition phase.
Table 16 shows the valid commands for selecting IN0 to IN15
or the temperature sensor. Commands other than those listed in
Table 16 are treated as no operation (NOOP) commands and
result in the multiplexer repeating the previous channel.
Single-cycle command mode minimizes the latency between
the 5-bit channel commands and the corresponding ADC data
because the multiplexer switches select the specified channel
immediately after the 5-bit command latches into memory. As
a result, the acquisition time depends on how quickly the digital
host can complete the write of the 5-bit command. Figure 95
shows a conversion mode timing diagram with single-cycle
command mode enabled, and Table 2 lists the relevant timing
specifications. The tACQ in single-cycle command mode is a
function of tCYC and the SCK period (tSCK), and can be calculated
with the following expression:
When two-cycle command mode is enabled, the first analog input
channel selected is the one specified in the AS_SLOT0 register. The
channel only updates when a valid command code is received.
To enable two-cycle command mode, set the STD_SEQ_EN bit to
0, set the NUM_SLOTS_AS field to 0x00 and set the
CYC_CTRL bit to 0 (see Table 13).
While the AD4695/AD4696 are in register configuration mode
when the STD_SEQ_EN bit in the SEQ_CTRL register is set to
0, the multiplexer automatically connects the channel specified
in the AS_SLOT0 register to the ADC core inputs, which allows
the ADC to acquire the signal on that channel even before the
device enters conversion mode.
t
ACQ = tCYC – tCONVERT – (5.5 × tSCK)
Table 16 shows the valid commands for selecting IN0 to IN15
or the temperature sensor. Commands other than those listed in
Table 16 are treated as NOOP commands and result in the
multiplexer repeating the previous channel.
When single-cycle command mode is enabled, the first analog
input channel selected is the one specified in the AS_SLOT0
register. The channel only updates after a valid command is
received.
Rev. 0 | Page 38 of 96
Data Sheet
AD4695/AD4696
To enable single-cycle command mode, set the STD_SEQ_EN bit
to 0, set the NUM_SLOTS_AS field to 0x00, and set the
CYC_CTRL bit to 1 (see Table 13).
mode enable settings, and threshold detection enable settings.
Configure each CONFIG_INn register before entering
conversion mode and initiating conversions.
When single-cycle command mode is enabled, the configuration
settings for each channel are set with the corresponding
CONFIG_INn register. Therefore, all analog inputs can have
different channel configuration options, analog input high-Z
Oversampling is not supported when single-cycle command
mode is enabled. Set the OSR for all analog inputs to 1 before
entering conversion mode with single-cycle command mode
enabled (see the Oversampling and Decimation section).
CNV
ADC
CONV
ACQ
CONV
ACQ
CONV
ACQ
CONV
ACQ
IN0
PHASE
MUX
CHANNEL
IN0
IN2
IN15
TEMP
BUSY
SDOx
IN0 RESULT
IN2 RESULT
IN15 RESULT
TEMP RESULT
Figure 73. Standard Sequencer Example with OSR = 1
CNV
ADC
PHASE
CONV
IN6
ACQ
CONV
ACQ
CONV
ACQ
CONV
ACQ
CONV
ACQ
IN6
MUX
CHANNEL
IN10
IN6
IN3
TEMP
BUSY
SDOx
IN6 RESULT
IN10 RESULT
IN6 RESULT
IN3 RESULT
TEMP RESULT
Figure 74. Advanced Sequencer Example with OSR = 1 for All Channels
CNV
1
2
N-1
N
1
ADC
PHASE
CONV
ACQ
CONV
ACQ
CONV
ACQ
CONV
ACQ
IN1
MUX
CHANNEL
IN0
BUSY
SDOx
IN0 RESULT
Figure 75. Standard Sequencer and Advanced Sequencer SPI Frames with IN0 OSR = N
CNV
ADC
PHASE
CONV
IN13
ACQ
CONV
ACQ
CONV
ACQ
CONV
ACQ
MUX
CHANNEL
IN9
IN8
TEMP
IN4
IN4
BUSY
ON GPx
SDI
CMD = 0x18
CMD = 0x0F
IN9 RESULT
CMD = 0x14
IN8 RESULT
CMD = 0x14
CMD = 0x1A
IN4 RESULT
SDOx
IN13 RESULT
TEMP RESULT
Figure 76. Two-Cycle Command Mode Timing
Rev. 0 | Page 39 of 96
AD4695/AD4696
Data Sheet
CNV
ADC
PHASE
CNV
ACQ CNV
TEMP
ACQ CNV
IN13
ACQ CNV
IN11
ACQ
MUX
CHANNEL
IN9
IN11
BUSY
ON GPx
SDI
CMD = 0x0F
CMD = 0x1D
CMD = 0x1B
CMD = NOOP
SDOx
IN9 RESULT
IN4 RESULT
IN13 RESULT
IN11 RESULT
Figure 77. Single-Cycle Command Mode Timing
Rev. 0 | Page 40 of 96
Data Sheet
AD4695/AD4696
DIGITAL INTERFACE
CS
SDI
The AD4695/AD4696 digital interface includes a 4-wire SPI, a
RESET
convert start input (CNV), an active low reset input ( ), and
ADDRESS
ADDRESS
WRITE DATA
PADDING
[CRC]
[CRC]
[CRC]
[CRC]
W
R
a BSY_ALT_GP0 pin that functions as a general-purpose pin.
WRITE
READ
SDO
SDI
The AD4695/AD4696 digital interface has two operating
modes, register configuration mode and conversion mode. In
register configuration mode, the SPI is used to read from and
write to the configuration registers. In conversion mode, the
SPI is used to read conversion results and optional status bits.
See the Register Configuration Mode section and Conversion
Mode section more details on these operating modes.
PADDING
SDO
READ DATA
INSTRUCTION PHASE
DATA PHASE
Figure 78. Basic SPI Frame
The interface logic level is set by the VIO voltage, and can range
from 1.2 V to 1.8 V. The AD4695/AD4696 use SPI Mode 3
(clock phase (CPHA) = clock polarity (CPOL) = 1).
Instruction Phase
Each SPI frame starts with the instruction phase. The instruction
CS
phase immediately follows a
falling edge (see Figure 78).
write
REGISTER CONFIGURATION MODE
W
The instruction phase consists of a read/
(R/ ) bit followed
by a register address word. Set the R/ bit high to initiate a read
W
When in register configuration mode, the digital host can read
from and write to the AD4695/AD4696 configuration registers
via the SPI. The device must be in register configuration mode to
perform register read and write instructions. Register configuration
mode is the default mode of operation on device power-up and
reset.
W
instruction or set the R/ bit low to initiate a write instruction.
The register address word specifies the address of the register to
be accessed. The register address word is 15 bits in length (long
addressing) by default, and can be changed to 7 bits in length
(short addressing) with the ADDR_LEN bit in the
SPI_CONFIG_B register.
The register configuration mode protocol is flexible and can be
configured for efficient access of large blocks of the configuration
register map. Each SPI frame consists of at least one instruction
phase, at least one data phase, and an optional 8-bit cyclic
redundancy check (CRC) checksum (see the Checksum
Protection section). Data is transmitted over the SPI MSB first.
The format and order of the instruction and data phases is
configurable, as described in the Instruction Phase section
through the Checksum Protection section. Figure 78 shows an
example of a basic SPI frame that consists of the instruction
phase, data phase, and optional CRC checksum.
When using single instruction mode, each register read or write
transaction in an SPI frame begins with an instruction phase.
When using streaming mode, only one instruction phase is
required per SPI frame to access a set of contiguous registers.
See the Single Instruction Mode section and the Streaming
Mode section for instructions on selecting and using these modes.
Data Phase
During the data phase, register data is either shifted out on
SDO on SCK falling edges (for register reads) or latched in on
SDI on SCK rising edges (for register writes). The data phase
can include the data for an entire register or individual bytes of
the register (see the Multibyte Register Access section).
CS
CS
rising
A
falling edge starts an SPI frame and a subsequent
edge ends the SPI frame. Data is latched on SDI on the SCK rising
edges and shifted out on SDO on the SCK falling edges. For all
SPI transactions, data is aligned MSB first.
If the CRC is disabled, the register contents are updated
immediately after the final SCK rising edge of the data phase. If
the CRC is enabled, the register contents are updated
immediately after the final SCK rising edge of the checksum (if
the checksum value matches the data in the data phase).
Figure 90 shows a detailed timing diagram for register read and
write operations via the SPI when the device is in register
configuration mode. See Table 2 for the timing specifications
shown in Figure 90.
Address Direction Options
See the Register Details section for a detailed description of the
addresses and functions of the AD4695/AD4696 configuration
registers.
The address direction options control whether the address is set
to automatically increment or decrement when accessing multiple
bytes of data in a single data phase (for example, when accessing
multibyte registers or when streaming mode is enabled). Figure 79
and Figure 80 show SPI frames with both address direction
options.
The 5-bit register configuration mode command switches the
device from conversion mode into register configuration mode
(see the Register Configuration Mode Command section).
Rev. 0 | Page 41 of 96
AD4695/AD4696
Data Sheet
Select between the two address direction options with the
ADDR_DIR bit in the SPI_CONFIG_A register. When the
ADDR_DIR bit is set to 0, the descending address option is
selected and the address decrements after each byte is accessed.
When the ADDR_DIR bit is set to 1, the ascending address
option is selected and the address increments after each byte is
accessed. The descending address option is selected by default.
individual bytes in a multibyte register (address = 0x0043) are
accessed over multiple SPI transactions in streaming mode and
single instruction mode with MB_STRICT = 1.
When the MB_STRICT bit is set to 1, all bytes of a multibyte
register must be read from or written to in the same SPI
transaction. With this setting, the data phase includes all bytes
when accessing a multibyte register. If the digital host fails to
read from or write to the entire multibyte register, the SPI
transaction is considered invalid and the MB_ERROR flag in the
SPI_STATUS register is set to 1. This setting ensures that all modes
or enable bits associated with a multibyte register are updated
simultaneously. The MB_STRICT bit is set to 1 by default.
Multibyte Register Access
Some AD4695/AD4696 configuration registers contain
multiple bytes of data stored in adjacent address locations in
memory. These registers are referred to as multibyte registers.
The address of each multibyte register is defined as the address
of its least significant byte (LSByte), but the multibyte register
contents extend across multiple register addresses. For example,
the STD_SEQ_CONFIG register (Address 0x024) is two bytes
long, the address of its LSByte is 0x024, and the address of its
MSByte is 0x025. Table 28 specifies whether registers are single
byte or multibyte
When the MB_STRICT bit is set to 1, the order in which each byte
of a multibyte register is read from or written to depends on the
selected address direction option (see the Address Direction
Options section). With the descending addresses option
selected, the first byte accessed in the data phase is the MSByte
of the multibyte register, and each subsequent byte corresponds
to the data in the next lowest address. With the ascending
addresses option selected, the first byte accessed in the data
phase is the LSByte of the multibyte register and each
subsequent byte corresponds to the data in the next highest
address. Figure 79 and Figure 80 show generalized read and
write transactions of a multibyte register for both address
direction options.
The state of the MB_STRICT bit in the SPI_CONFIG_C
register determines whether multibyte registers are treated as a
single unit of memory with one register address or as multiple
registers that are each one byte long with individual register
addresses.
When the MB_STRICT bit is set to 0, each byte of a multibyte
register must be read from or written to individually, which
allows the digital host to access one byte of a multibyte register
without accessing the other byte(s). With this setting, all data
phases in an SPI frame consist of a single byte rather than the
entire multibyte register, and each byte in a multibyte register is
directly addressable. The contents of either byte are updated by
an SPI write transaction as long as new data is provided for that
entire byte. Figure 82 and Figure 87 show examples where
When CRC is enabled, a checksum follows the data phase for
each SPI transaction. When the MB_STRICT bit is set to 0, the
checksum occurs after each byte of a multibyte register is
accessed (see Figure 82 and Figure 87). When the MB_STRICT
bit is set to 1, the checksum only occurs after all bytes of the
multibyte register are accessed (see Figure 83 and Figure 88).
CS
SDI
SDO
SDI
MSBYTE ADDRESS MSBYTE DATA LSBYTE DATA
[CRC]
[CRC]
[CRC]
[CRC]
W
R
WRITE
READ
PADDING
PADDING
MSBYTE ADDRESS
SDO
MSBYTE DATA LSBYTE DATA
INSTRUCTION PHASE
DATA PHASE
Figure 79. Multibyte Register Access with MB_STRICT = 1 and Descending Address
CS
SDI
SDO
SDI
LSBYTE ADDRESS LSBYTE DATA MSBYTE DATA
PADDING
[CRC]
[CRC]
[CRC]
[CRC]
W
R
WRITE
READ
LSBYTE ADDRESS
PADDING
SDO
LSBYTE DATA MSBYTE DATA
INSTRUCTION PHASE
DATA PHASE
Figure 80. Multibyte Register Access with MB_STRICT = 1 and Ascending Address
Rev. 0 | Page 42 of 96
Data Sheet
AD4695/AD4696
address decrements until it reaches Address 0x0000 and the
Streaming Mode
address is set to the highest valued register address available
(Address 0x013F) on the subsequent byte access. If looping is
disabled and the ascending address option is selected, the
address increments until it reaches the highest valued register
address available (Address 0x013F), and the address is set to
Address 0x0000 on the subsequent byte access. Looping is
disabled by default.
When the INST_MODE bit in the SPI_CONFIG_B register is
set to 0, streaming mode is enabled. In streaming mode, only
one instruction phase is required per SPI frame and the register
address being read from or written to is automatically updated
after each data phase (based on the selected address direction
option). The instruction phase is followed by multiple data
phases for each register being accessed until the end of the SPI
frame. Streaming mode enables efficient access to large,
contiguous sections of the configuration register map, such as
when updating the advanced sequencer slot registers
Note that even when using 7-bit addressing, registers with
addresses larger than 0xFF are still accessible in streaming
mode. However, accessing these registers is generally more
efficient using 15-bit addressing.
(AS_SLOTn) to configure the advanced sequencer.
Single Instruction Mode
Figure 81 shows a generalized SPI frame for performing
multiple register read and write transactions with streaming
mode selected. Because there is only one instruction phase per
frame in streaming mode, all SPI transactions in a given SPI
frame are either all reads or all writes. The checksum is
included in each data phase only if CRC is enabled (see the
Checksum Protection section).
When the INST_MODE bit in the SPI_CONFIG_B register is
set to 1, single instruction mode is enabled. In single
instruction mode, each SPI read or write transaction includes
an instruction phase to specify whether the transaction is a read
or a write and what address is being accessed. Single instruction
mode allows the digital host to quickly read from and write to
registers with nonadjacent register addresses in a single SPI
frame, as opposed to streaming mode, which allows exclusively
reading from or writing to registers with adjacent addresses
without starting a new SPI frame.
Figure 82 to Figure 84 show examples of accessing different
parts of the register map with both address direction options
and with both MB_STRICT options (see the Multibyte Register
Access section).
Figure 86 shows a generalized SPI frame for performing multiple
register read and write transactions with single instruction
mode selected. The checksum is included in each data phase
only if CRC is enabled (see the Checksum Protection section).
When streaming mode is active, a specified number of registers
can be looped to repeatedly access the same registers multiple
times in a single SPI frame. The LOOP_COUNT field in the
LOOP_MODE register determines how many registers are
accessed before the register address is reset to the starting
address (the one specified in the instruction phase). When the
MB_STRICT bit is set to 1, a multibyte register is considered
one register when looping. When the MB_STRICT bit is set to
0, each byte of a multibyte register is considered one register
when looping. Figure 85 shows an example using looping to
repeatedly read from the CLAMP_STATUSn registers.
Figure 87 shows an example of reading from and writing to the
MSByte and LSByte of the UPPER_IN1 register (MB_STRICT =
0). Figure 88 and Figure 89 show examples of reading from the
UPPER_IN1 register and writing to the UPPER_IN0 register in the
same frame with both address direction options (MB_STRICT =
1). Note that the UPPER_INn registers are multibyte registers,
and when MB_STRICT is set to 1, both bytes must be read
from or written to in one data phase (see the Multibyte Register
Access register section).
If LOOP_COUNT is set to 0x0, looping is disabled. If looping is
disabled and the descending address option is selected, the
CS
SDI
SDO
SDI
ADDRESS
ADDRESS
DATA
PADDING
PADDING
DATA
[CRC]
[CRC]
[CRC]
[CRC]
DATA
[CRC]
[CRC]
DATA
[CRC]
[CRC]
W
R
WRITE
READ
PADDING
PADDING
PADDING
PADDING
SDO
DATA
[CRC]
DATA
[CRC]
INSTRUCTION PHASE
DATA PHASE
DATA PHASE
DATA PHASE
Figure 81. Streaming Mode SPI Frame
Rev. 0 | Page 43 of 96
AD4695/AD4696
Data Sheet
0x0043
0x0042
0x0041
CS
SDI
ADDR = 0x0043 DATA[7:0] [CRC]
DATA[7:0] [CRC]
DATA[7:0] [CRC]
W
R
WRITE
SDO
SDI
PADDING
[CRC]
[CRC]
PADDING
[CRC]
PADDING
[CRC]
ADDR = 0x0043 PADDING
PADDING
READ
SDO
DATA[7:0] [CRC]
DATA PHASE
DATA[7:0] [CRC]
DATA PHASE
DATA[7:0] [CRC]
DATA PHASE
INSTRUCTION PHASE
Figure 82. Streaming Mode SPI Frame, Looping Disabled, Descending Address, MB_STRICT = 0
0x0043
0x0041
0x003F
CS
SDI
ADDR = 0x0043
ADDR = 0x0043
DATA[15:0] [CRC] DATA[15:0] [CRC]
DATA[7:0]
PADDING
[CRC]
[CRC]
W
R
WRITE
READ
SDO
SDI
PADDING
PADDING
[CRC]
[CRC]
PADDING
[CRC]
PADDING
SDO
DATA[15:0] [CRC] DATA[15:0] [CRC]
DATA[7:0]
[CRC]
INSTRUCTION PHASE
DATA PHASE
DATA PHASE
DATA PHASE
Figure 83. Streaming Mode SPI Frame, Looping Disabled, Descending Address, MB_STRICT = 1
0x003F
0x0040
0x0042
CS
SDI
ADDR = 0x003F
ADDR = 0x003F
DATA[7:0] [CRC] DATA[15:0] [CRC] DATA[15:0] [CRC]
W
R
WRITE
READ
SDO
SDI
PADDING
PADDING
[CRC]
[CRC]
PADDING
[CRC]
PADDING
[CRC]
PADDING
SDO
DATA[7:0] [CRC] DATA[15:0] [CRC] DATA[15:0] [CRC]
INSTRUCTION PHASE
DATA PHASE
DATA PHASE
DATA PHASE
Figure 84. Streaming Mode SPI Frame, Looping Disabled, Ascending Address, MB_STRICT = 1
Figure 85. Streaming Mode SPI Frame, Looping Enabled, LOOP_COUNT = 7, Descending Address
Rev. 0 | Page 44 of 96
Data Sheet
AD4695/AD4696
Figure 86. Single Instruction Mode SPI Frame
CS
SDI
R
ADDR = 0x0043
PADDING
[CRC]
ADDR = 0x0042
DATA[7:0] [CRC]
W
SDO
DATA[7:0] [CRC]
DATA PHASE
PADDING
[CRC]
INSTRUCTION PHASE
INSTRUCTION PHASE
DATA PHASE
Figure 87. Single Instruction Mode SPI Frame, MB_STRICT = 0
CS
SDI
R
ADDR = 0x0043
PADDING
[CRC]
ADDR = 0x0041
DATA[15:0] [CRC]
W
SDO
DATA[15:0] [CRC]
DATA PHASE
PADDING
[CRC]
INSTRUCTION PHASE
INSTRUCTION PHASE
DATA PHASE
Figure 88. Single Instruction Mode SPI Frame, MB_STRICT = 1, Descending Address
CS
SDI
R
ADDR = 0x0042
PADDING
[CRC]
ADDR = 0x0040
DATA[15:0] [CRC]
W
SDO
DATA[15:0] [CRC]
DATA PHASE
PADDING
[CRC]
INSTRUCTION PHASE
INSTRUCTION PHASE
DATA PHASE
Figure 89. Single Instruction Mode SPI Frame, MB_STRICT = 1, Ascending Address
performing multiple register reads with streaming mode
Checksum Protection
selected, where the digital host is only required to send a CRC
on SDI for the first transaction (see Figure 81).
The AD4695/AD4696 include optional error checking based on
an 8-bit CRC in register configuration mode. When the CRC is
enabled, an 8-bit checksum code is appended to the data phase
of each register read or write transaction. The value of the
checksum is calculated from the data read or written over the
SPI, and therefore allows the AD4695/AD4696 and the digital
host to detect corrupted data. If the checksum does not match
the corresponding register data, the register read or write is
considered invalid.
When the AD4695/AD4696 receive a checksum that does not
match its corresponding SPI transaction, the transaction is
considered invalid, and the CRC_ERROR bit in the SPI_
status register is set to 1. The CRC_ERROR bit is a write 1 to
clear bit (R/W1C) and must be written to 1 to be cleared.
When a write transaction is considered invalid, register contents
are not updated. When a read transaction is considered invalid,
the digital host must ignore the received register data and
attempt the register read transaction again. Read to clear bits
are also not cleared unless the register read transaction is
considered valid (for example, the HI_INn and LO_INn bits in
the ALERT_STATUSn registers).
Figure 81 shows a generalized SPI frame for performing register
reads and writes with streaming mode selected, including the
CRC checksum. Figure 86 shows a generalized SPI frame for
performing register reads and writes with single instruction
mode selected, including the CRC checksum. Note that the
checksums on SDI shown in both Figure 81 and Figure 86 are
sent from the digital host to the AD4695/AD4696, and the
digital host must send a valid checksum during the SPI read
and write transactions pictured. The only exception is when
When streaming mode and the CRC are both enabled and an
invalid checksum is received for a given SPI transaction, all
subsequent SPI transactions are considered invalid for the
CS
remainder of the SPI frame (until
is brought high).
Rev. 0 | Page 45 of 96
AD4695/AD4696
Data Sheet
The CRC is enabled with the CRC_EN and CRC_EN_N fields
in the SPI_CONFIG_C register. To enable the CRC, CRC_EN
must be set to 0x1 and CRC_EN_N must be set to 0x2. CRC is
disabled for all other combinations of CRC_EN and
CRC_EN_N.
Each SPI transaction has a corresponding code and the
polynomial is applied to that code to generate a checksum. The
codes consist of an 8-bit seed value appended to data from the
SPI transaction. Table 14 shows the data and seed values for
each possible type of SPI transaction.
The AD4695/AD4696 expect checksums to be included in each
SPI transaction immediately after the CRC is enabled. Write to
the SPI_CONFIG_C register to enable the CRC before writing
to any other registers, then read the SPI_CONFIG_C register
assuming that the CRC has been enabled. If the SPI master
receives the correct state of the CRC_EN and CRC_EN_N
fields and a valid checksum, the CRC is enabled, and the SPI
master can begin configuring the remaining configuration
registers.
In single instruction mode, the seed for all CRCs is 0xA5. In
streaming mode, the seed for the first CRC in the frame is also
0xA5, but the seed for the remaining CRCs in the frame is the
LSByte of the register address being accessed. If MB_STRICT is
set to 1 and a multibyte register is accessed, the register address
used for the seed depends on the selected address direction
option. The address of the MSByte is used with descending
address, and the address of the LSByte is used with ascending
address. For example, in both Figure 83 and Figure 84, the
second data phase includes data from the UPPER_IN0 register,
but the seed used for the checksum is 0x41 with the descending
address option (Figure 83) and 0x40 with the ascending address
option (Figure 84).
The following CRC polynomial is used to calculate the
checksums:
x8 + x2 + x + 1
Table 14. CRC Input Values for SPI Modes and Transactions
SPI Transaction Type
Pin
Single Instruction Mode or First CRC with Streaming Mode
Subsequent CRCs with Streaming Mode
SPI data = data phase bits
Seed = LSByte of current register address
SPI data = data phase bits
Seed = LSByte of current register address
Not applicable
Write
SDI
SPI data = instruction phase bits, data phase bits
Seed = 0xA5
SDO SPI data = instruction phase bits, data phase bits
Seed = 0xA5
SDI
Read
SPI data = instruction phase bits, padding bits
Seed = 0xA5
SDO SPI data = instruction phase bits, data phase bits
Seed = 0xA5
SPI data = data phase bits
Seed = LSByte of current register address
Rev. 0 | Page 46 of 96
Data Sheet
AD4695/AD4696
The length of the data phase (represented by N in Figure 90)
depends on whether the CRC is enabled and the length of the
register being accessed (see the Checksum Protection and
Multibyte Register Access sections).
Register Read and Write Timing Diagrams
Figure 90 shows a timing diagram for the SPI when the AD4695/
AD4696 are in register configuration mode. See Table 2 for the
timing specifications pictured in Figure 90.
The AD4695/AD4696 ignore the state of CNV when in register
configuration mode. The Entering Conversion Mode section
describes the process for placing the AD4695/AD4696 in
conversion mode.
CS
Register read and write transactions are framed by . While
CS
is high, SCK edges are ignored, and SDO is high impedance.
CS
A falling edge on
begins an SPI frame and data on SDI is
latched on SCK rising edges while data is shifted out on SDO
Entering Conversion Mode
CS
on SCK falling edges. A rising edge on
and forces SDO to high impedance.
ends the SPI frame
To place the AD4695/AD4696 in conversion mode, set
the SPI_MODE bit in the setup register to 1. When the
SPI_MODE bit is set to 1, the SPI frame immediately terminates,
and the device enters conversion mode. No further register
reads or writes can occur until the device enters register
configuration mode again.
CS
The first phase of an SPI frame immediately following a
falling edge is the instruction phase. The instruction phase is
followed by the data phase. For SPI read transactions, the
register contents are shifted out on SDO during the data phase.
For SPI write transactions, the register contents are latched in
on SDI during the data phase. See the Streaming Mode and
Single Instruction Mode sections for a detailed description of
the order of instruction and data phases in each SPI frame.
The digital host must provide a delay specified by tSCKCNV after
the final SCK rising edge of the register write before initiating
conversions with a CNV rising edge (see Table 2 and Figure 90).
The length of the address in the instruction phase (represented
by M in Figure 90) is set by the ADDR_LEN bit in the
SPI_CONFIG_B register (see the Instruction Phase section).
tCNVL
CNV
tCSBH
tSCKCNV
CS
tEN
tSCK
tSCKH
tSCKL
tSCKCSB
SCK
tHSDI
tSSDI
SDI
ADDR[M – 1]
ADDR[M – 2]
DATA[N – 1]
DATA[0]
CSBDIS
R/W
CSBDIS
DSDO
HSDO
HIGH-Z
SDO
DATA[N – 1]
DATA[0]
Figure 90. Register Configuration Mode Timing Diagram
Rev. 0 | Page 47 of 96
AD4695/AD4696
Data Sheet
When autocycle mode is enabled, the AD4695/AD4696
CONVERSION MODE
generate their own internal convert start signal to autonomously
perform conversions without a CNV signal from the digital
host (see the Autocycle Mode section).
When the AD4695/AD4696 are in conversion mode, CNV
rising edges initiate conversions on the selected channel and
the channel sequencing logic updates the multiplexer to the
next channel (see the Converter Operation and Channel
Sequencing Modes sections). The device enters conversion
mode when the SPI_MODE bit in the setup register is set to 1.
Status Bits
A set of five status bits can be appended to the end of each
conversion result. The status bits allow the digital host to
monitor the status of the analog inputs without interrupting
analog-to-digital conversions. Table 15 shows the names and
descriptions of the status bits.
In conversion mode, the SPI is used to read the ADC results and
write the 5-bit SDI commands shown in Table 16. Figure 91 to
Figure 96 show timing diagrams for SPI frames relative to
CS
performing conversions. The CNV pin and
pin can be tied
By default, the OV_ALT status bit indicates the status of the
overvoltage clamp flags (the bitwise logical OR of the
CLAMP_FLAG bit and COM_CLAMP_FLAG bit in the status
register). When the OV_ALT_MODE bit in the GP_MODE
register is set to 1, the OV_ALT status bit is the logical OR of
the CLAMP_FLAG bit and the threshold detection alert
indicator (TD_ALERT bit in the status register). The digital
host can monitor the state of the OV_ALT bit to detect and
respond to out of range events.
together to enable interfacing with a single 4-wire SPI port (see
Figure 96). Each ADC result is available until the next CNV
rising edge occurs.
An optional set of five status bits can be appended to the ADC
data. The status bits include channel information, the
overvoltage clamp flag, and a threshold detection alert
indicator. See the Status Bits section for a description of the
status bits and how they are enabled.
The INX bits indicate which of the 16 analog inputs the
conversion result corresponds to. The values for the INX bits
range from 0 to 15 (0x0 to 0xF) and correspond to IN0 to IN15,
respectively. An INX value of 15 corresponds to either IN15 or
the temperature sensor. The INX bits can be used by the digital
host to align the ADC data with the sequence of analog input
channels.
In conversion mode, the BSY_ALT_GP0 pin can be assigned as
an additional serial data output to reduce the SCK frequency
required to shift out the ADC result plus optional status bits
before the next conversion occurs. See the Serial Data Output
Modes section for a description of the options available on both
package options of the AD4695/AD4696 and how to enable
these modes.
Set the STATUS_EN bit in the setup register to 1 to enable the
status bits. The status bits are disabled by default. When the
status bits are enabled, the serial data output word extends to 24
bits, where Bit 20 to Bit 24 contain the status bits (see Table 19 and
Table 20).
The BSY_ALT_GP0 pin can also be assigned as either the busy
indicator or the threshold detection alert indicator. Figure 91 to
Figure 96 show the relative timing of the CNV signal and the
busy and alert indicators when they are assigned to
BSY_ALT_GP0. The General-Purpose Pin section describes
how to set BSY_ALT_GP0 to the desired function.
Table 15. Status Bits Names and Descriptions
Status Word Index
Bit Name
Description
Bit 4
OV_ALT
Active high. Indicates the status of the overvoltage protection clamp flag and (if enabled) the status
of the threshold detection alert indicator.
Bits[3:0]
INX
Indicates what analog input channel the ADC data corresponds to (IN0 to IN15).
Table 16. Conversion Mode Commands
Channel Sequencing Mode
5-Bit SDI Command (CMD)
Description
Two-Cycle Command Mode and Single-Cycle Command Mode
0x00 to 0x09, 0x0B to 0x0E
NOOP
0x0A
0x0F
0x10 to 0x1F
Register configuration mode command
Temperature sensor channel selection
IN0 to IN15 channel selection
NOOP
Standard Sequencer and Advanced Sequencer
0x00 to 0x09, 0x0B to 0x1F
0x0A
Register configuration mode command
Rev. 0 | Page 48 of 96
Data Sheet
AD4695/AD4696
The serial data output modes only apply when the device is in
conversion mode. In register configuration mode, register read
data is always shifted out serially on SDO only.
Serial Data Output Modes
The AD4695/AD4696 digital interface allows clocking out ADC
data on more than one serial data output, which reduces the
number of SCK periods required to access the full ADC result
and allows slower SCK frequencies. The two serial data output
modes include single-SDO mode and dual-SDO mode. In single-
SDO mode, the ADC results are only shifted out on SDO. In
dual-SDO mode, ADC results are shifted out on SDO and
BSY_ALT_GP0 in parallel.
The SDO_MODE field in the setup register determines which
serial data output mode is selected. Table 18 shows the values of
SDO_MODE and the corresponding serial data output modes.
Table 17. Serial Data Output Mode Pin Assignments
Mode
Signal
SDO0
SDO1
SDO0
Pin
Single-SDO Mode
Dual-SDO Mode
SDO
BSY_ALT_GP0
SDO
Table 17 shows the pins used for each serial data output signal for
each serial data output mode. Table 19 and Table 20, show the
formatting of conversion results for all combinations of serial
data output modes, status bits, and OSR options. The values of
the blank cells in Table 19 and Table 20 depend on the setting
of SDO_STATE, as described in the Conversion Mode Timing
Diagrams section.
Table 18. SDO_MODE Values vs. Serial Data Output Mode
SDO_MODE
Mode
0x0
0x1
0x2
0x3
Single-SDO Mode
Dual-SDO Mode
Single-SDO Mode
Single-SDO Mode
Table 19. Single-SDO Mode Data Output Format
SCK Falling Edge Number
15 16 17 18 19 20
D1 D0
D2 D1 D0
D3 D2 D1 D0
OSR Setting Status Bits Signal
1
2
3
…
…
…
…
…
…
…
…
…
21
22
23
24
1
Disabled
Disabled
Disabled
Disabled
Enabled
Enabled
Enabled
Enabled
SDO0
SDO0
SDO0
SDO0
SDO0
SDO0
SDO0
SDO0
D15 D14 D13
D16 D15 D14
D17 D16 D15
D18 D17 D16
D15 D14 D15
D16 D15 D16
D17 D16 D17
D18 D17 D16
4
16
64
1
D4 D3 D2 D1 D0
D1 D0
D2 D1 D0
D3 D2 D1 D0
0
0
0
0
0
0
OV_ALT INX[3] INX[2] INX[1] INX[0]
OV_ALT INX[3] INX[2] INX[1] INX[0]
OV_ALT INX[3] INX[2] INX[1] INX[0]
4
16
64
D4 D3 D2 D1 D0 OV_ALT INX[3] INX[2] INX[1] INX[0]
Table 20. Dual-SDO Mode Data Output Format
SCK Falling Edge Number
OSR Setting
Status Bits
Signal
SDO1
SDO0
SDO1
SDO0
SDO1
SDO0
SDO1
SDO0
SDO1
SDO0
SDO1
SDO0
SDO1
SDO0
SDO1
SDO0
1
2
3
4
5
6
7
8
9
10
11
12
1
Disabled
D15
D14
D16
D15
D17
D16
D18
D17
D15
D14
D16
D15
D17
D16
D18
D17
D13
D12
D14
D13
D15
D14
D16
D15
D13
D12
D14
D13
D15
D14
D16
D15
D11
D10
D12
D11
D13
D12
D14
D13
D11
D10
D12
D11
D13
D12
D14
D13
D9
D8
D10
D9
D11
D10
D12
D11
D9
D8
D10
D9
D7
D6
D8
D7
D9
D8
D10
D9
D7
D6
D8
D7
D9
D8
D10
D9
D5
D4
D6
D5
D7
D6
D8
D7
D5
D4
D6
D5
D7
D6
D8
D7
D3
D2
D4
D3
D5
D4
D6
D5
D3
D2
D4
D3
D5
D4
D6
D5
D1
D0
D2
D1
D3
D2
D4
D3
D1
D0
D2
D1
D3
D2
D4
D3
4
Disabled
Disabled
Disabled
Enabled
Enabled
Enabled
Enabled
D0
16
64
1
D1
D0
D2
D1
0
0
D0
0
D1
D0
D2
D1
D0
0
INX[3]
INX[1]
INX[3]
INX[1]
INX[3]
INX[1]
INX[3]
INX[1]
INX[2]
INX[0]
INX[2]
INX[0]
INX[2]
INX[0]
INX[2]
INX[0]
OV_ALT
0
OV_ALT
0
OV_ALT
D0
OV_ALT
4
16
64
D11
D10
D12
D11
Rev. 0 | Page 49 of 96
AD4695/AD4696
Data Sheet
The SDO_STATE bit in the setup register determines the
Conversion Mode Timing Diagrams
behavior of the serial data output(s) at the beginning and the end
of the conversion mode SPI frames. When the SDO_STATE bit is
set to 0, the serial data output(s) hold their final value(s) until
the MSB of the next conversion result is clocked out. The serial
data output(s) remain in this state even if multiple extra SCK
falling edges occur after the full result is shifted out. The serial
Figure 91 to Figure 96 show detailed timing diagrams for
performing analog-to-digital conversions when the AD4695/
AD4696 are in conversion mode with each serial data output
mode option (with autocycle mode disabled).
When the device is in conversion mode, a CNV rising edge
initiates a conversion and enters the conversion phase (see the
Converter Operation section). When a conversion is initiated, it
continues until completion regardless of the state of CNV. When
the standard sequencer, advanced sequencer, or two-cycle
command mode is enabled, the device enters the acquisition
phase before the conversion phase is complete. When single-
cycle command mode is enabled, the device enters the
acquisition phase after the sixth SCK rising edge in the SPI
frame. Figure 91 through Figure 94 and Figure 96 show tACQ
when the standard sequencer, advanced sequencer, or two-
cycle command mode is enabled. Figure 95 shows tACQ when
single-cycle command mode is enabled.
CS
data output(s) are forced to high impedance when
is brought
CS
high, but return to the previous state after
is brought low
again. Figure 91 and Figure 93 show the behavior of the serial
data output(s) when SDO_STATE is set to 0. SDO_STATE is
set to 0 by default.
When SDO_STATE is set to 1, the busy indicator is enabled on
the serial data output(s) (see the Busy Indicator section). The
serial data output(s) are forced to high impedance if any SCK
falling edges occur after the final bits of the result are already
CS
clocked out, or when CNV or
is brought high. When a CNV
rising edge initiates a conversion, the serial data output(s) remain
high impedance until the conversion phase is complete and the
result is available to be read over the SPI. The serial data output(s)
are driven low when the data is ready. If the current selected
channel has an OSR greater than 1, the serial data output(s) are
CS
CS
is high, SCK edges
frames the conversion result data. While
are ignored, and all pins assigned as serial data outputs are high
CS
impedance. While
is low, data is clocked out with the MSB first
on the serial data output(s) on SCK falling edges, and data is
latched in on SDI on the SCK rising edges.
CS
driven low after the oversampled result is ready. Note that
must be driven low for the busy indicator to appear on the
serial data output(s).
CS
CNV and
can be tied together and driven by the chip select
of the SPI master to minimize the number of digital signals
required to interface with the AD4695/AD4696 (see the SPI
Peripheral Connections section). Figure 96 shows a timing
diagram of the AD4695/AD4696 interfacing with a 4-wire SPI
When the busy indicator is enabled on BSY_ALT_GP0, the
BSY_ALT_GP0 pin is driven high after a CNV rising edge and
is driven low when the conversion is complete (see the Busy
Indicator on BSY_ALT_GP0 section). The signal labeled busy
in Figure 91 to Figure 95 represents the BSY_ALT_GP0 pin
assigned as the busy indicator. Figure 75 in the Channel
Sequencing Modes section shows the relative timing of the CNV
rising edge and the busy indicator on BSY_ALT_GP0 for OSR
settings of 1 and greater than 1.
CS
with the CNV and
signals tied together.
The conversion phase must be complete before the digital host
provides the first SCK falling edge. The digital host can use the
busy indicator falling edge to detect the end of the conversion
phase and to begin clocking out the ADC results. Otherwise,
the digital host must include a delay dictated by the conversion
time specification (tCONVERT) in Table 2 between the CNV rising
edge and the first SCK falling edge.
When the threshold detection alert indicator is enabled on
BSY_ALT_GP0, the BSY_ALT_GP0 pin reflects the value of the
TD_ALERT bit in the status register. The signal labeled alert in
Figure 91 to Figure 95 represents the BSY_ALT_GP0 assigned
as the alert indicator. Figure 75 in the Channel Sequencing
Modes section show the relative timing of the CNV rising edge
and the alert indicator on the BSY_ALT_GP0 pin for OSR
settings of 1 and greater than 1.
The 5-bit SDI commands shown in Table 16 are latched in on
SDI on the first five SCK rising edges in the SPI frame. The register
configuration mode command instructs the AD4695/AD4696 to
exit conversion mode and enter register configuration mode
(see the Register Configuration Mode Command section). The
channel select commands in Table 16 are only used when two-
cycle command mode or single-cycle command mode is
enabled, and are interpreted as NOOP commands when the
standard sequencer or the advanced sequencer is enabled (see
the Channel Sequencing Modes section).
Register Configuration Mode Command
The register configuration mode command is a 5-bit command
written on SDI that instructs the device to exit conversion mode
and enter register configuration mode. The register configuration
mode command is 0x0A. Figure 97 shows the relative timing of
the register configuration mode command and the AD4695/
AD4696 entering register configuration mode.
To ensure optimal performance, there must be a sufficient
delay between the final SCK edge and the next CNV rising edge,
and there must be no SCK activity until the conversion time has
elapsed (see tSCKCNV in Table 2 and Figure 91 to Figure 96).
Rev. 0 | Page 50 of 96
Data Sheet
AD4695/AD4696
The register configuration mode command is clocked in on SDI on
the first five SCK rising edges after a conversion. When the register
configuration mode command is received, the subsequent rising
mode. The digital host must wait for the tREGCONFIG delay (shown in
Figure 97 and Table 2) to elapse between the fifth SCK rising edge
CS
and the
rising edge.
CS
edge on places the AD4695/AD4696 in register configuration
CYC
tCONVERT
CNVH
ACQ
CNVL
CNV
ADC
CONVERSION
PHASE
ACQUISITION
CNVBSY
SCKCNV
BUSY
CNVALT
ALERT
CSBH
CS
SCKCSB
EN
SCK
SCKH
4
SCKL
15 TO
23
16 TO
24
SCK
1
2
3
5
6
HSDI
SSDI
SDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
DSDO
CMD[0]
DON'T CARE
CSBDIS
HSDO
CSBDIS
LSB
SDO LSB
MSB
MSB – 1
MSB – 2
MSB – 3
MSB – 4
LSB+1
LSB
N
N–1
N–1
N
N
N
N
N
N
Figure 91. Conversion Mode Timing Diagram, Single-SDO Mode, SDO_STATE = 0
Rev. 0 | Page 51 of 96
AD4695/AD4696
Data Sheet
CYC
tCONVERT
CNVH
ACQ
CNVL
CNV
ADC
PHASE
CONVERSION
ACQUISITION
CNVBSY
SCKCNV
BUSY
CNVALT
ALERT
CSBH
CS
SCKCSB
EN
SCK
SCKH
4
SCKL
15 TO
23
16 TO
24
1
2
3
5
6
SCK
HSDI
SSDI
SDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
DSDO
CMD[0]
DON'T CARE
CSBDIS
HSDO
CSBDIS
SDO
MSB
MSB – 1
MSB – 2
MSB – 3
MSB – 4
LSB + 1
LSB
Figure 92. Conversion Mode Timing Diagram, Single-SDO Mode, SDO_STATE = 1
CYC
tCONVERT
CNVH
ACQ
CNVL
CNV
ADC
CONVERSION
CNVBSY
ACQUISITION
PHASE
SCKCNV
BUSY
CNVALT
CSBH
ALERT
CS
SCKCSB
EN
SCK
SCKH
4
SCKL
7 TO
11
8 TO
12
1
2
3
5
6
SCK
HSDI
SSDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
CMD[0]
DON'T CARE
SDI
DSDO
HSDO
CSBDIS
LSB–1
LSB–1
LSB
MSB
MSB – 2
MSB – 4
MSB – 6
MSB – 8
MSB – 9
LSB + 3
LSB + 1
N
N–1
N–1
N
N
N
N
N
N
N
SDO1
SDO0 LSB
MSB – 1
N
MSB – 3
MSB – 5
MSB – 7
LSB + 2
LSB
N
N–1
N
–1
N
N
N
N
Figure 93. Conversion Mode Timing Diagram, Dual-SDO Mode, SDO_STATE = 0
Rev. 0 | Page 52 of 96
Data Sheet
AD4695/AD4696
CYC
tCONVERT
CNVH
ACQ
CNVL
CNV
ADC
PHASE
CONVERSION
CNVBSY
ACQUISITION
SCKCNV
BUSY
CNVALT
CSBH
ALERT
EN
SCK
SCKH
4
SCKL
7 TO
11
8 TO
12
6
1
2
3
5
SCK
HSDI
SSDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
DSDO
CMD[0]
DON'T CARE
HSDO
CSBDIS
CSBDIS
SDO1
SDO0
MSB
MSB – 2
MSB – 3
MSB – 4
MSB – 5
MSB – 6
MSB – 8
MSB – 9
LSB + 3
LSB + 2
MSB – 1
MSB – 7
Figure 94. Conversion Mode Timing Diagram, Dual-SDO Mode, SDO_STATE = 1
CYC
tCONVERT
ACQ
CNVH
CNVL
CNV
ADC
PHASE
CONVERSION
CNVBSY
ACQUISITION
SCKCNV
BUSY
CSBH
CS
EN
SCKCSB
SCK
SCKH
4
SCKL
15 TO
23
16 TO
24
1
2
3
5
6
SCK
HSDI
SSDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
CMD[0]
DON'T CARE
SDI
Figure 95. Conversion Mode Timing Diagram, Single-Cycle Command Mode Enabled
Rev. 0 | Page 53 of 96
AD4695/AD4696
Data Sheet
CYC
tCONVERT
CNVH
ACQ
CNV
CS
ADC
CONVERSION
PHASE
ACQUISITION
SCKH
SCK
SCKL
EN
SCKCNV
SCKCSB
15 TO
23
16 TO
24
SCK
SDI
1
2
3
4
5
6
HSDI
SSDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
DSDO
CMD[0]
DON'T CARE
LSB + 1
HSDO
CSBDIS
CSBDIS
SDO
MSB
MSB – 1
MSB – 2
MSB – 3
MSB – 4
LSB
Figure 96. Conversion Mode Timing Diagram with 4-Wire SPI, Single-SDO Mode, SDO_STATE = 1
CNVH
CNV
CONVERT
BUSY
CSBH
REGCONFIG
CS
EN
SCK
SCKH
4
SCKL
SCKCSB
16 TO
24
1
2
3
5
6
7
SCK
HSDI
SSDI
SDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
CMD[0]
DON'T CARE
CSBDIS
DSDO
HSDO
CSBDIS
LSB
LSB
MSB
MSB – 1
MSB – 2
MSB – 3
MSB – 4
MSB – 5
LSB
N
SDO
N-1
N-1
N
N
N
N
N
N
Figure 97. Conversion Mode Timing Diagram, Register Configuration Mode Command
AUTOCYCLE MODE
Table 21. Autocycle Mode Conversion Period Options
The AD4695/AD4696 can be configured to convert autonomously
on a user-programmed channel sequence, which is the ideal mode
of operation for system monitoring. When autocycle mode is
enabled, the AD4695/AD4696 generate an internal clock that
acts as the convert start signal, and the digital host is not
required to generate a signal on CNV. The internal convert
start clock is enabled when the AD4695/AD4696 enter
conversion mode. The internal convert start clock is disabled
when the AD4695/AD4696 enter register configuration mode.
Therefore, conversions only occur when the AD4695/AD4696
are in conversion mode.
AC_CYC, Bits[2:0]
Value
Conversion Period
(μs)
Sample Rate
(kSPS)
0x0
0x1
0x2
0x3
0x4
0x5
0x6
0x7
10
20
40
80
100
200
400
800
100
50
25
12.5
10
5
2.5
1.25
Autocycle mode can be used in conjunction with the busy
indicator, threshold detection alerts, and the standard or
advanced sequencers to reduce overhead for the digital host
system. The threshold detection alert indicator can be assigned
to the BSY_ALT_GP0 pin and used as an interrupt to indicate a
predetermined out of bounds event. The threshold detection
interrupt service routine can optionally trigger an SPI
Autocycle mode is enabled when the AC_EN bit in the
AC_CTRL register is set to 1. There are eight options for the
period of the internal convert start signal. The convert start
signal period is selected with the AC_CYC field in the AC_CTRL
register. Table 21 shows the conversion period and
corresponding sample rates for each AC_CYC value.
instruction to read back the most recent conversion result and
Rev. 0 | Page 54 of 96
Data Sheet
AD4695/AD4696
exit conversion mode to determine the specific type of out of
bounds event using the alert indicator registers
(ALERT_STATUS1 to ALERT_STATUS4).
transaction before the next conversion begins (see the Conversion
Mode SPI Clock Frequency Requirements section).
The tACBSY specification dictates how long the busy indicator is
low between two conversions when autocycle mode is enabled.
The tSCKCNV specification dictates how much time must be given
between the final SCK rising edge of the SPI transaction and the
start of the next conversion.
Note that the SPI transactions when autocycle mode is enabled
must adhere to the timing specifications of conversion mode
(see the Conversion Mode section and Table 2). Either the alert
indicator or the busy indicator can be assigned to the
BSY_ALT_GP0 pin to determine when the digital host can
initiate the SPI transaction. See the General-Purpose Pin
section for a description of configuring the BSY_ALT_GP0 pin
to output the busy indicator or the alert indicator.
The tCNVALT specification indicates the delay between the start of
the conversion and when the alert indicator state is updated. A
rising edge of the alert indicator does not directly imply that the
AD4695/AD4696 interface is ready for an SPI transaction, but
can be used as an interrupt to trigger an SPI transaction if the
transaction is completed before the remainder of tCYC elapses.
As shown in Figure 98, SPI transactions when using autocycle
mode must not start before tCONVERT has elapsed. The busy
indicator or the alert indicator must be used to ensure the digital
host is synchronized to the internal convert start clock (see the SPI
Peripheral Synchronization in Autocycle Mode section). The SCK
rate must also be fast enough to complete the desired SPI
Autocycle mode is intended to be used with the standard
sequencer and advanced sequencer to minimize the overhead
for the digital host. Autocycle mode can be used with two-cycle
command mode and single-cycle command mode, but the digital
host must transmit the 5-bit SDI commands for selecting channels.
CYC
tCONVERT
ACQ
ADC
CONVERSION
PHASE
ACQUISITION
ACBSY
BUSY
CNVALT
ALERT
CSBH
SCKCNV
CS
SCKCSB
EN
SCK
SCKH
4
SCKL
15 TO
23
16 TO
24
SCK
SDI
1
2
3
5
6
HSDI
SSDI
CMD[4]
CMD[3]
CMD[2]
CMD[1]
CMD[0]
DON'T CARE
CSBDIS
DSDO
HSDO
CSBDIS
LSB
LSB
MSB
MSB – 1
MSB – 2
MSB – 3
MSB – 4
LSB + 1
LSB
N
N–1
SDO
N–1
N
N
N
N
N
N
Figure 98. Conversion Mode Timing Diagram with Autocycle Mode Enabled (Single-SDO Mode, SDO_STATE = 0)
Rev. 0 | Page 55 of 96
AD4695/AD4696
Data Sheet
the GPIO_STATE register to set the state of this signal. Bit 0 of
the GPO_WRITE field in the GPIO_STATE register controls the
state of the BSY_ALT_GP0 pin when configured as a general-
purpose output (see Table 52). The logic output thresholds for
GENERAL-PURPOSE PIN
Table 22 shows the functions available on the BSY_ALT_GP0 pin,
plus the function priority (lower numbers indicate higher priority).
To configure the BSY_ALT_GP0 for a given function, all higher
priority functions must be disabled. The Busy Indicator section,
Threshold Detection and Alert Indicators section, Serial Data
Output Modes section, and GPIO section describe the behavior
of the BSY_ALT_GP0 when the BSY_ALT_GP0 pin is
the BSY_ALT_GP0 pin are specified in Table 1 as VOL and VOH
.
DEVICE RESET
A device reset reinitializes the AD4695/AD4696 configuration
registers. The AD4695/AD4696 provide several options for
performing a device reset, including a hardware reset, a
software reset, and PORs.
configured for each function shown in Table 22.
When the BSY_ALT_GP0 pin is configured for any function
other than a general-purpose input, the BSY_ALT_GP0 pin
functions as a digital output. If another device attempts to drive
the BSY_ALT_GP0 pin while the BSY_ALT_GP0 pin is
configured as a digital output, contention occurs and can
damage the AD4695/AD4696. The BSY_ALT_GP0 pin is
configured as a digital input by default.
Hardware resets, software resets, and PORs all assert the
RESET_FLAG bit in the status register. The RESET_FLAG bit
is a read to clear bit and is automatically set to 0 after a valid
read from the status register. The RESET_FLAG bit can be used
by the digital host to confirm that the device has executed a
device reset, or if a reset was performed unintentionally.
Table 22. General-Purpose Pin Functions and Function
Priority of the BSY_ALT_GP0 Pin
Function Priority
All device reset methods require a delay between the start of the
reset instruction and when the AD4695/AD4696 SPI is ready to
receive communications from the digital host. The device reset
delays are shown in Figure 99 through Figure 106 and in Table 2.
When the digital host attempts to perform an SPI read or write
transaction before the device is ready, the transaction is
considered invalid and the NOT_RDY_ERROR bit in the
SPI_STATUS is set to 1. The NOT_RDY_ERROR bit is a
R/W1C bit and is only reset when set to 1 with a valid register
write transaction.
1 (Highest
Priority)
4 (Lowest
Priority)
Pin
2
3
BSY_ALT_GP0 SDO1
signal
Alert
Busy
GPIO
indicator indicator
(dual-SDO
mode)
GPIO
The BSY_ALT_GP0 pin can be configured as a general-purpose
input or general-purpose output using the GPI0_EN and
GPO0_EN bits in the GPIO_CTRL register, respectively (see
Table 50). The BSY_ALT_GP0 bit is configured as an input
when the corresponding GPI0_EN bit is set to 1 and is
configured as an output when the corresponding GPO0_EN bit
is set to 1.
Hardware Reset
RESET
A hardware reset is initiated by the
shows a timing diagram for performing a hardware reset. tRESETL is
RESET
falling edge. Figure 99
the minimum amount of time that
and tHWR_DELAY is the time that the digital host must wait between a
RESET
must be driven low,
falling edge and starting an SPI frame (see Table 2).
If the internal LDO supplies VDD, and the internal LDO is
disabled before a hardware reset, the internal LDO is enabled by
the hardware reset and an additional delay is required to account
for the internal LDO output reaching the VDD minimum required
voltage (see the Power-On Resets section).
The GPIO functionality on the BSY_ALT_GP0 pin allows the
digital host to monitor logic outputs or control logic inputs of
other devices in the system through the AD4695/AD4696 SPI
instead of using additional digital host GPIO pins. The
AD4695/AD4696 GPIO functionality is especially useful in
digitally isolated applications because the functions reduce the
number of required digital isolation channels.
Software Reset
To initiate a software reset, set the SW_RST_MSB bit and
SW_RST_LSB bit in the SPI_CONFIG_A register to 1. A software
reset reinitializes the state of all configuration registers listed in
the Register Information section to the default values, except
for the SPI_CONFIG_A register. When the software reset is
complete, the SW_RST_MSB bit and SW_RST_LSB bit
automatically clear. Figure 100 shows the timing requirements
for performing a software reset. tSWR_DELAY is the time that the
digital host must wait between the software reset and starting a
new SPI frame (see Table 2).
When the BSY_ALT_GP0 pin is configured as a general-
purpose input, the BSY_ALT_GP0 pin can be connected to a
logic output of another device in the system and the digital host
can read the GPIO_STATE register to monitor the
BSY_ALT_GP0 pin state. Bit 0 of the GPI_READ field in the
GPIO_STATE register indicates the state of the BSY_ALT_GP0
pin (see Table 52). The logic input thresholds for the
BSY_ALT_GP0 pin are specified in Table 1 as VIL and VIH.
When the BSY_ALT_GP0 pin is configured as a general-purpose
output, the BSY_ALT_GP0 pin can be connected to a logic input
of another device in the system, such as other multiplexers or
programmable gain amplifiers, and the digital host can write to
Rev. 0 | Page 56 of 96
Data Sheet
AD4695/AD4696
Figure 104 shows a timing diagram of a VIO POR where the
internal LDO is used to supply VDD. tPOR_VIO2 is the time that the
digital host must wait between VIO being supplied and starting an
SPI frame.
Power-On Resets (PORs)
A POR is initiated when VDD or VIO is first supplied. When a
POR event is detected, the AD4695/AD4696 configuration
registers are initialized to the default values, but it is still
recommended to perform either a hardware reset or a software
reset after a POR.
When the internal LDO supplies VDD, a POR occurs when the
internal LDO is enabled by the LDO wake-up command or by a
hardware reset if the internal LDO was previously disabled
(LDO_EN bit = 0). Figure 105 shows a timing diagram of a POR
where the internal LDO is enabled by the LDO wake-up
command. tWAKEUP_SW is the time that the digital host must wait
between the LDO wake-up command and starting a new SPI
frame. Figure 106 shows a timing diagram of a POR where the
internal LDO is enabled by a hardware reset. tWAKEUP_HW is the
time that the digital host must wait between the hardware reset
and starting an SPI frame.
Figure 101 shows a timing diagram of a VDD POR where VIO is
already supplied. tPOR_VDD is the time that the digital host must wait
between VDD first being supplied and starting an SPI frame (see
Table 2). Figure 102 shows a timing diagram of a VIO POR where
VDD is already supplied. tPOR_VIO1 is the time that the digital host
must wait between VIO first being supplied and starting an SPI
frame (see Table 2).
When VDD is supplied by the internal LDO, the VDD POR is
triggered when the internal LDO output drives VDD to at least the
minimum VDD specification. The internal LDO output is only
enabled when both LDO_IN and VIO are supplied, and when the
LDO_EN bit in the SETUP register is set to 1 (see the Internal
LDO section).
tPOR_LDO, tPOR_VIO2, tWAKEUP_HW, and tWAKEUP_SW all depend on the VDD
decoupling capacitance (CVDD). Larger values of CVDD increase the
amount of time it takes for the internal LDO output voltage to
reach the minimum VDD supply voltage to trigger a VDD POR.
Table 2 provides typical values for these reset delay specifications
with CVDD = 1 μF.
Figure 103 shows a timing diagram of an LDO_IN POR where
VIO is already supplied. tPOR_LDO is the time that the digital host
must wait between LDO_IN first being supplied and starting an
SPI frame.
VDD
RESET
CS
SCK
SDI
DON'T CARE
REG READ/WRITE DATA
Figure 99. Hardware Reset Timing Diagram
VDD
CS
CSBH
SWR_DELAY
SCK
SDI
SW RESET INSTRUCTION
DON'T CARE
REG READ/WRITE DATA
Figure 100. Software Reset Timing Diagram
Rev. 0 | Page 57 of 96
AD4695/AD4696
Data Sheet
VIO
LDO_IN
VDD
CS
SCK
SDI
DON'T CARE
REG READ/WRITE DATA
Figure 101. VDD POR Timing Diagram
VIO
POR_VIO1
VDD
CSBH
CS
SCK
SDI
Figure 102. VIO POR Timing Diagram (VDD Supplied Externally)
VIO
LDO_IN
VDD
POR_LDO
CSBH
CS
SCK
SDI
DON'T CARE
REG READ/WRITE DATA
Figure 103. LDO_IN POR Timing Diagram (Internal LDO Supplying VDD)
VIO
LDO_IN
POR_VIO2
VDD
CS
SCK
SDI
DON'T CARE
REG READ/WRITE DATA
Figure 104. VIO POR Timing Diagram (Internal LDO Supplying VDD)
Rev. 0 | Page 58 of 96
Data Sheet
AD4695/AD4696
VIO
LDO_IN
VDD
CSBH
CS
WAKEUP_SW
SCK
SDI
LDO WAKEUP COMMAND
DON'T CARE
REG READ/WRITE DATA
Figure 105. LDO Wake-Up Command POR Timing Diagram
LDO_IN
VDD
RESET
CS
SCK
SDI
Figure 106. POR with Internal LDO Enabled by Hardware Reset Timing Diagram
Rev. 0 | Page 59 of 96
AD4695/AD4696
Data Sheet
APPLICATIONS INFORMATION
Figure 107 shows an example of the recommended connection
diagram for the AD4695/AD4696 companion circuitry.
optional digital isolation). The following sections provide
recommendations and suggestions for selecting and connecting
the AD4695/AD4696 companion circuitry based on common
application requirements.
The AD4695/AD4696 companion circuitry includes power
supplies, voltage reference circuitry, analog front-end signal
conditioning, and an SPI-compatible digital controller (plus
+VS
ADR4550,
ADR435
REFERENCE
CIRCUITRY
LDO
LDO
LDO
1
1µF
100nF
100nF
VLOGIC
AVDD LDO_IN VDD
VIO
C
REF
REF
IN0
CNV
CLOCK
+VS
R
ADA4807-x,
ADA4805-x
EXT,0
CS
SCK
SDI
CS
C
EXT,0
SCLK
MOSI
MISO0
VIN
–VS
+VS
0
DIGITAL
HOST
AD4695/AD4696
SDO
R
EXT,15
ADA4807-x,
ADA4805-x
IN15
C
EXT,15
BSY_ALT_GP0
RESET
MISO1, GPIOx, IRQ
2
VIN
–VS
+VS
15
GPIO
R
EXT,COM
ADA4807-x,
ADA4805-x
GND
COM
C
REFGND
AGND
IOGND
EXT,COM
REFGND
OR
–VS
V
/2
REF
3
1. DEDICATED +1.8V SUPPLY FOR VDD NOT REQUIRED WHEN USING INTERNAL LDO
2. R AND C REPRESENT EXTERNAL INPUT RC FILTERS FOR ANALOG INPUT INx OR COM
EXT,X
EXT,X
3. DEDICATED ADC DRIVER FOR COM INPUT NOT REQUIRED IF CONNECTING COM TO REFGND
Figure 107. Typical Connection Diagram
Rev. 0 | Page 60 of 96
Data Sheet
AD4695/AD4696
At the start of the conversion phase, the multiplexer switches
are disconnected and the voltage on the currently selected
analog input channel is sampled on the capacitive DAC. During
the acquisition phase, the multiplexer switches (SWMUX+ and
SWMUX−) close to connect the next selected analog input
channel to the capacitive DAC. A voltage glitch (commonly
referred to as kickback) occurs when these switches close due to
the difference between the voltage on the capacitive DAC and
the voltage on the selected analog input pins.
ANALOG FRONT-END DESIGN
The analog front-end companion circuitry for the
AD4695/AD4696 normally includes an external RC filter and
an ADC driver or precision operational amplifier between the
signal being measured and the AD4695/AD4696 analog inputs.
The component selection and design of the analog front-end
circuitry driving the AD4695/AD4696 analog inputs have a
direct impact on overall system performance. The analog front-
end must be designed with the system target noise, accuracy,
distortion and settling requirements of the end application. The
following sections provide recommendations for designing
analog front-end and signal conditioning circuits based on
these requirements.
To achieve specified performance of the AD4695/AD4696, this
kickback must be settled to within half an LSB of the ADC core
before the start of the next conversion phase (that is, the next
CNV rising edge). The rate at which the kickback voltage is
settled depends on the transient characteristics and bandwidth
of the analog front-end circuitry. Signal settling requirements
therefore dictate the minimum allowable analog front-end
bandwidth and constrain the driver amplifier and external RC
filter selection.
External RC Filter
The external RC low-pass filter consists of an external resistor
and capacitor (represented by REXT and CEXT in Figure 64 and
Figure 107). These components act to reduce the wideband
noise from the analog front-end circuitry, reduce the nonlinear
voltage kickback that occur at the analog inputs, and protect
the analog inputs from overvoltage events. Selecting the
appropriate values of REXT and CEXT for these functions is
described in the Analog Front-End Noise Considerations section,
the Signal Settling section, and the Analog Input Overvoltage
Protection section.
Table 23 provides a list of recommended amplifiers and
external RC filter components for various sample rates and
signal bandwidths. Figure 69 and Figure 70 in the Analog Input
High-Z Mode section show SNR and THD performance with
various amplifiers and external RC component values.
Analog input high-Z mode significantly reduces the bandwidth
requirements of the analog front end by minimizing the size of
the voltage kickback. Figure 20 shows the difference in
magnitude of the kickback when analog input high-Z mode is
disabled and enabled.
Ensure that the CEXT capacitor is an NP0 ceramic capacitor to
limit distortion artifacts, and that the PCB layout minimizes the
parasitic impedance between CEXT and the analog input pin. See
the Layout Guidelines section for more information.
Analog Front-End Noise Considerations
Signal Settling Requirements
The magnitude of the analog front-end noise directly impacts
the dynamic range and SNR performance of the overall AD4695/
AD4696 signal chain. Select the analog front-end components
and configuration to achieve the target noise specification for
the overall system.
As described in the Converter Operation and Analog Inputs
sections, the AD4695/AD4696 analog inputs (IN0 to IN15 and
COM) are routed to the ADC core inputs via the internal
analog multiplexer.
As shown in Figure 64, the ADC core capacitive DAC can be
represented by a switched capacitive load.
Figure 108 illustrates the primary noise sources in a typical
analog front-end driver circuit.
R
F
V
=
4kTR
n_RF
F
V
n
R
R
V
=
=
G
4kTR
4kTR
V
n_RG
G
n_ADC
R
EXT
V
i
+
–
n_total
n–
+
n_AFE
–
C
V
EXT
S
V
n_RS
S
i
n+
Figure 108. Noise Sources in Typical ADC Analog Front-End Circuit
Rev. 0 | Page 61 of 96
AD4695/AD4696
Data Sheet
2
2
Assuming all noise sources are Gaussian and uncorrelated, the
total system rms noise (vn_total) is calculated as follows:
2
2
RF
RF
2
4kTRF 1
4kTRs i Rs n
4kTRG i RF
n
n
RG
RG
2
2
where:
νn_total
where:
n_AFE is the referred to output (RTO) rms noise of the analog
front end.
n_ADC is the AD4695/AD4696 input referred rms noise.
=
n _ AFE n _ ADC
k is the Boltzmann constant.
T is the absolute temperature in Kelvin.
RF and RG are the feedback network resistors, as shown in
Figure 108.
v
v
RS is the source resistance, as shown in Figure 108.
i
n+ and in− represent the amplifier input current noise spectral
The estimated system dynamic range (DRtotal) is a measure of
the system rms noise and the full-scale input range.
density in pA/√Hz.
vn represents the amplifier input voltage noise spectral density
in nV/√Hz.
VREF / 2 2
DRtotal = 20log
n _total
See MT-049 and MT-050 for detailed derivations of vn_AFE vs.
analog front-end components and configurations.
The AD4695/AD4696 input referred rms noise specification
(vn_ADC) is typically 37.8 μV rms (see Table 1). Figure 109 shows
Analog Front-End Noise in Pseudo Bipolar Mode
When configuring a channel in pseudo bipolar mode, typically
a second analog front-end circuit is required to drive the
negative-side input to VREF/2 V (as described in the Channel
Configuration Options section). In this case, the RTO rms
noise of the additional analog front end (vn_AFE2) is added to the
rss equation to calculate the total system rms noise:
the typical system dynamic range vs. vn_AFE with vn_ADC
=
37.8 μV rms and VREF = 5 V. For vn_AFE less than 13 μV rms, the
overall system dynamic range remains within 0.5 dB of the
AD4695/AD4696 dynamic range specification (see Table 1).
95
2
2
2
90
85
80
75
70
νn_total
=
n _ AFE n _ AFE2 n _ ADC
Note that the bandwidth of the RC filter and values of REXT and
CEXT cannot be set arbitrarily low due to the settling requirements
of the AD4695/AD4696 analog inputs. Refer to the Signal
Settling Requirements section for guidelines on selecting the
optimal RC filter components for the target sample rate.
Guidelines for Driver Amplifier Selection
The following is a list of guidelines for selecting the amplifier(s)
used in the AD4695/AD4696 analog front-end based on the
end system requirements.
1
10
V
100
500
(µV rms)
n_AFE
The amplifier voltage and current noise specifications must be
sufficiently low to achieve the desired rms noise and dynamic
range performance, as described in the Analog Front-End
Noise Considerations section.
Figure 109. AD4695/AD4696 Typical Dynamic Range vs. vn_AFE, VREF = 5 V
The analog front-end RTO noise (vn_AFE) is equal to the rms
noise of each of the constituent components in the analog
front-end, referred to the output of the external RC filter (REXT and
CEXT in Figure 107 and in the External RC Filter section).
Assuming the RC filter bandwidth is much lower than the
bandwidth of the amplifier circuit, vn_AFE is equal to the noise
spectral density of each of these components (referred to the
amplifier output) multiplied by the effective noise bandwidth of
the RC filter (ENBWRC), where:
The distortion performance of the amplifier must be sufficient
to achieve desired THD performance. To meet the
AD4695/AD4696 THD data sheet specification, the amplifier
circuit must have lower or comparable distortion specifications.
The small signal bandwidth of the amplifier must be sufficiently
higher than the minimum bandwidth required to adequately
settle the voltage steps that occur when switching between two
analog input channels, as described in the Signal Settling
section.
2
1
ENBWRC
=
2 REXTCEXT
and
The amplifier should also have sufficient supply headroom to
adequately output a full-scale signal to the AD4695/AD4696
analog inputs (see the input voltage range specification in Table 1).
Refer to the input and output headroom requirements in the
amplifier data sheet to determine the supply voltages required
to support the desired full-scale range for the given channel.
νn_AFE = ENBWRC
×
Rev. 0 | Page 62 of 96
Data Sheet
AD4695/AD4696
The ADA4805-1 and ADA4807-1 and their dual- and quad-
amplifier models are suitable amplifiers for channels acquiring
ac waveforms, due to their exceptionally low noise and
distortion and high bandwidth.
precision. It is recommended to enable analog input high-Z
mode on AD4695/AD4696 analog input channels when being
driven directly by the ADA4610-1, ADA4077-1, or amplifiers
with similar bandwidth specifications to ensure adequate
settling performance (see the Signal Settling Requirements
section and Analog Input High-Z Mode section).
The ADA4610-1 and ADA4077-1 and their dual- and quad-
amplifier models are suitable amplifiers for channels
monitoring dc or low frequency signals that require high
Table 23. Recommended Amplifier and External RC Filter Component Selection Recommendations
Input Signal Bandwidth (kHz)
Sample Rate
Amplifier
REXT (Ω)
390
390
680
680
680
680
680
680
200
200
200
390
390
390
CEXT (pF)
180
180
180
180
180
180
470
470
180
180
180
180
180
180
≤10
≤1 MSPS
ADA4805-1/ADA4805-2
ADA4807-1/ADA4807-2/ADA4807-4
ADA4610-1/ADA4610-2/ADA4610-4
ADA4077-1/ADA4077-2/ADA4077-4
ADA4805-1/ADA4805-2
ADA4807-1/ADA4807-2/ADA4807-4
ADA4610-1/ADA4610-2/ADA4610-4
ADA4077-1/ADA4077-2/ADA4077-4
ADA4805-1/ADA4805-2
ADA4807-1/ADA4807-2/ADA4807-4
ADA4896-2
ADA4805-1/ADA4805-2
ADA4807-1/ADA4807-2/ADA4807-4
ADA4896-2
≤500 kSPS
>10
≤1 MSPS
≤500 kSPS
Rev. 0 | Page 63 of 96
AD4695/AD4696
Data Sheet
voltage that can be supported for a given analog input is a
ANALOG INPUT OVERVOLTAGE PROTECTION
function of VREF and REXT. The following relation can be used to
determine the required value of REXT to limit the clamp current
to the maximum supported current (5 mA) given the VREF and
maximum expected vIN voltage:
The external resistor in the external RC filter (represented by
REXT in Figure 64, Figure 107, and Figure 110) works with the
input overvoltage protection clamps to provide overvoltage
protection to the analog inputs (see the Input Overvoltage
Protection Clamps section).
R
EXT + νIN,max – VREF/5 mA (ꢀ)
For example, if the analog input source can swing to 7.5 V and
VREF = 5 V, REXT must be approximately 500 Ω to limit the
clamp current to 5 mA. If this resistor is being sized based upon
the clamping current limits, CEXT must be carefully chosen to
ensure adequate input bandwidth is achieved (see the Analog
Front-End Noise Considerations and Signal Settling sections
for more information).
An overvoltage event is defined as an event where the overvoltage
protection clamps are activated as a result of the input voltage
on IN0 to IN15 or COM exceeding the clamp activation voltage
specification (VACT in Figure 110). The maximum VACT voltage
specification is VREF + 0.55 V (see Table 1).
When activated, the clamp on the given channel sinks current
from the source to ground (see ICLAMP in Figure 110), resulting
in a voltage drop across REXT. The AD4695/AD4696 overvoltage
protection clamps support a maximum sustained ICLAMP current
of 5 mA (see Table 1). REXT therefore isolates the analog input
pin voltage from the applied voltage (vIN). The maximum vIN
The value of REXT also must be selected to ensure stability of the
overvoltage protection clamp circuit, if desired. See the
Overvoltage Protection Clamp Stability section for more
information.
I
CLAMP
R
EXT
INx, COM
+
+
≥V
C
ACT
EXT
–
I
V
CLAMP
CLAMP
–
AD4695/AD4696
Figure 110. Analog Input Overvoltage Event
Rev. 0 | Page 64 of 96
Data Sheet
AD4695/AD4696
is required if the selected voltage reference does not have an
adequate load regulation to drive the REF input at the desired
ADC sample rate (see the Reference Circuit Design for Driving
REF Input section).
REFERENCE CIRCUITRY DESIGN
The AD4695/AD4696 VREF sets the full-scale range of the ADC
core and determines the resulting output code for a given
analog input voltage (see the Transfer Function section). The
VREF voltage therefore has a direct impact on the overall system
accuracy and ac performance. The reference companion circuitry
for the AD4695/AD4696 must have adequate noise
performance, accuracy, drift and signal settling characteristics
for the end application.
CREF supplies the charge necessary for the ADC core to perform
the bit trials as part of the conversion phase and filters noise
from the other reference circuitry. CREF must be sufficiently
large to prevent deviations in VREF during the ADC bit trials.
When reference input high-Z mode is enabled, the amount of
charge pulled by the REF input is significantly reduced, thereby
reducing the minimum CREF capacitance. When reference input
high-Z mode is disabled, a 10 μF CREF is recommended. When
reference input high-Z mode is enabled, a 1 μF CREF is
recommended.
The REF input is a dynamic current load that pulls charge from
the reference circuitry during the conversion phase of the ADC
core, and the reference circuit must be able to maintain a stable
VREF while the ADC is performing conversions to maintain
performance (that is, gain error).
The PCB layout of the reference circuitry relative to the AD4695/
AD4696 REF input is critical to ensuring optimal performance.
The Layout Guidelines section provides recommendations and
guidelines for the layout of the reference circuit components.
The AD4695/AD4696 reference input high-Z mode significantly
reduces the magnitude of the average current of the REF input
when enabled. The reference input high-Z mode significantly
reduces the drive requirements of the reference circuitry,
allowing system designers to prioritize dc accuracy, power, and
system footprint targets.
REFERENCE CIRCUIT DESIGN FOR DRIVING REF
INPUT
Figure 111 shoes the typical connection diagram for the AD4695/
AD4696 companion reference circuit. The reference circuitry
consists of a voltage reference, CREF, and any accompanying
reference buffer or analog low-pass filtering. A reference buffer
Figure 111 shows a typical connection diagram for the
reference circuitry driving the REF input of the
AD4695/AD4696.
OPTIONAL EXTERNAL
REFERENCE BUFFER
+V
ADR4550,
ADR435
VOLTAGE
+V
ADA4807
REFERENCE
OPTIONAL LOW-PASS
2
FILTER
C
REF
V
REF
1
REFIN
REF
AD4695/AD4696
1
WHEN DRIVING REF DIRECTLY ON WLCSP OPTION, SHORT REFIN TO REF AND ENSURE THE
INTERNAL REFERENCE BUFFER AND REFERENCE BUFFER BYPASS OPTION ARE DISABLED.
(REFBUF_EN = REFBUF_BP = 0)
2
ADDITIONAL LOW-PASS FILTERING MUST NOT BE IMPLEMENTED WITHOUT A REFERENCE BUFFER.
Figure 111. Typical Connection Diagram for Driving REF Input
Rev. 0 | Page 65 of 96
AD4695/AD4696
Data Sheet
The device driving the REF input must have sufficiently low
output impedance so that the reference input current does not
cause VREF to deviate enough to violate the system performance
targets. To achieve data sheet performance, VREF must remain
within half an LSB. The maximum output impedance of the
device driving the REF input (Ro_max) is therefore:
where Lmax is the load regulation specification for the voltage
reference in ppm/mA that corresponds to the calculated Ro_max
.
IREF is typically 11 μA at 1 MSPS with reference input high-Z mode
enabled and 320 μA at 1 MSPS with reference input high-Z mode
disabled (in unipolar mode). IREF scales linearly with the ADC
sample rate (see Table 1 and Figure 39). The output impedance
and load regulation requirements of the reference circuitry are
therefore less strict at lower sample rates. Table 24 shows
calculated Ro_max and Lmax for VREF = 5 V and for different
sample rates and with reference input high-Z mode disabled
and enabled. Table 24 also provides recommendations for
voltage references and discrete reference buffers for each of
these conditions.
16 1
VREF / 2
Ro_max
=
IREF
where IREF is the average REF input current.
Most voltage references specify load regulation in ppm/mA,
which can be converted to effective output impedance with the
following:
Ro _ max
Lmax = 1000 ×
VREF
Table 24. Reference Circuitry Recommendations, REF Input
Reference Input
Sample Rate High-Z Mode
IREF (ꢀA) Ro_max (Ω) Lmax (ppm/mA) Recommended Voltage References and Reference Buffers
1 MSPS
1 MSPS
500 kSPS
500 kSPS
Disabled
Enabled
Disabled
Enabled
320
12
160
6
0.12
3.2
0.24
6.4
24
640
48
ADR4550 with ADA4807-1, ADR445 with ADA4807-1, ADR435
ADR4550, ADR445, ADR435
ADR445 with ADA4807-1, ADR4550, ADR435
ADR4550, ADR445, ADR435
320
Rev. 0 | Page 66 of 96
Data Sheet
AD4695/AD4696
and the effective noise remains vn_total. When OSR settings of 4,
16, and 64 are used, the noise is attenuated by a factor of 2, 4,
and 8, respectively.
CONVERTING BETWEEN CODES AND VOLTS
The Transfer Function section describes the ideal transfer
function between the analog input voltage sampled by the
AD4695/AD4696 ADC core and the resulting output code. The
analog input voltage (VINx) corresponding to each possible
output code value (CODEOUT) is a function of the VREF voltage
and the OSR setting and polarity mode for the selected channel:
The resulting dynamic range when utilizing oversampling
(DROSR) is as follows:
DROSR = DRtotal + 10 log(OSR)
where DRtotal is the system dynamic range for an OSR of 1
(defined in the Analog Front-End Noise Considerations
section).
VREF
V
INx = LSB × CODEOUT
=
× CODEOUT
2N
where:
LSB is the LSB size.
N is the resolution of the output code. The AD4695/AD4696
ADC core outputs 16-bit results (N = 16), but the output code
resolution is a function of the OSR selected for the given
channel (DR):
The effective number of bits (ENOB) of the system increases by
1 every time the noise is halved. As a result, ENOB increases by
1 bit every time the OSR is increased by a factor of 4. To reflect
this, when an AD4695/AD4696 channel is configured with OSR
settings of 4, 16, or 64, the resolution of the conversion results
for that channel is extended to 17 bits, 18 bits, and 19 bits,
respectively (see the Transfer Function section and Serial Data
Output Modes section).
N = 16 log4(OSR)
OSR can be set to 1, 4, 16, or 64, which correspond to an output
code resolution of 16, 17, 18, and 19, respectively. Table 7
through Table 10 show the negative and positive full-scale
output code values for each OSR. See the Oversampling and
Decimation section for details on configuring the OSR for each
channel.
Note that oversampling and decimation only reduce voltage
noise for uniformly distributed Gaussian noise sources and
have no effect on other types of noise sources (such as 1/f
noise).
DIGITAL INTERFACE OPERATION
The polarity mode for the selected channel determines whether
CODEOUT uses straight binary or twos complement format.
When unipolar mode is selected, CODEOUT is in straight binary,
and is therefore an unsigned integer value. When pseudo
bipolar mode is selected, CODEOUT uses twos complement
encoding, and is therefore a signed integer value. See the
Channel Configuration Options for details on configuring the
polarity mode for each channel.
Figure 107 shows a typical connection diagram of the
AD4695/AD4696 digital interface connected to a digital host.
The AD4695/AD4696 can be operated by a single 4-wire SPI-
compatible host, but some features require additional digital
resources, such as GPIOs and timers.
The following sections provide recommendations for digital
interface connections and operation to interact with the
AD4695/AD4696 interface and feature set.
The offset and gain correction settings for each channel modify
the transfer function of the AD4695/AD4696 to correct for
first-order inaccuracies in the system that cause the observed
transfer function to deviate from the ideal. Update the offset
and gain fields for each channel during system calibration. The
Offset and Gain Correction section describes how the offset
and gain fields modify the AD4695/AD4696 transfer function.
ADC Convert Start Signal Options
The CNV input is analogous to an edge triggered interrupt pin,
which instructs the AD4695/AD4696 ADC core to perform a
conversion (see the Converter Operation section). The CNV
input is active only when the AD4695/AD4696 are in
conversion mode and is ignored when in register configuration
mode. The period of the signal driving the CNV input sets the
sample rate of the AD4695/AD4696, and must conform to the
tCYC specification in Table 2 and in Figure 91 through Figure 95.
OVERSAMPLING FOR NOISE REDUCTION
The AD4695/AD4696 include on-chip oversampling and
decimation as a means to reduce the total effective Gaussian
noise of the system in the digital domain (see the Oversampling
and Decimation section). Assuming the analog front-end noise
is Gaussian, the effective system noise after oversampling
(vn_OSR) is:
The ADC core samples the analog input voltage on the selected
channel on the rising edge of CNV. The signal driving the CNV
input therefore must have sufficiently low jitter and fast edge
rates to achieve the desired noise performance at the target input
frequencies. The layout of the trace connecting the AD4695/
AD4696 CNV input to the digital host must be as short as
possible with minimal vias to minimize trace impedance (see
the Layout Guidelines section).
vn _ total
vn _OSR
OSR
where OSR is the oversampling ratio setting for the given
analog input channel and vn_total is the RTO system noise
(defined in the Analog Front-End Noise Considerations
section). When the OSR is set to 1, no oversampling occurs,
In conversion mode, the digital host SPI master peripheral
must be synchronous to the CNV signal and follow the timing
requirements specified in the Conversion Mode Timing
Diagrams section. See the SPI Peripheral Synchronization in
Rev. 0 | Page 67 of 96
AD4695/AD4696
Data Sheet
Conversion Mode section for recommendations for
maintaining proper SPI timing.
line, especially when the busy indicator is enabled on SDO (see
the Busy Indicator on Serial Data Outputs section).
Embedded clock divider or timer peripherals can typically
output an integer division of the system clock. When utilizing
embedded clock divider peripherals, connect the digital host
clock output to CNV and set the clock output frequency to the
desired sample rate. The clock output must be enabled while
the AD4695/AD4696 are in conversion mode, but it can be
either enabled or disabled while in register configuration mode.
CS
SCK
SDI
CS
SCLK
MOSI
VIO
AD4695/AD4696
SPI MASTER
≥ 2 kΩ
(SDO0)
SDO
MISO0
MISO1
(SDO1)
1
BSY_ALT_GP0
CS
The CNV input can be connected to the
output of the SPI
CS
master peripheral, provided the
rising edge timing is
1
BSY_ALT_GP0 FUNCTIONS AS SDO1 WHEN DUAL-SDO MODE IS ENABLED.
GP1 FUNCTIONS AS SDO1.
deterministic and periodic (see Figure 114). Note that for OSR
settings greater than 1, multiple CNV rising edges are required
before the result is available to be read out on the SPI (see the
Oversampling and Decimation section). The SPI outputs all 0s
Figure 113. AD4695/AD4696 SPI Connection Diagram (Dual-SDO Mode)
Figure 113 shows a connection diagram for interfacing the
AD4695/AD4696 SPI to a digital host SPI master peripheral
when configured in dual-SDO mode. Route BSY_ALT_GP0 to
the second MISO input on the digital host (MISO1). It is
recommended to include pull-up resistors on both SDO0 and
SDO1 lines, especially when the busy indicator is enabled on
the serial data outputs (see the Busy Indicator on Serial Data
Outputs section).
CS
during the CNV/ frames prior to the data being ready.
An external crystal oscillator with a CMOS clock driver can also
drive the CNV input. With this option, either the oscillator
output or the busy indicator from the AD4695/AD4696 must
be routed to the digital host and used as a timer or interrupt trigger
to achieve synchronization between the CNV signal and the SPI
master peripheral (see the SPI Peripheral Synchronization in
Conversion Mode section).
SPI Peripheral Synchronization in Conversion Mode
The AD4695/AD4696 have a 4-wire SPI in SPI Mode 3 for
accessing register contents and ADC results. The digital host
must at minimum include a 4-wire SPI-compatible peripheral
to operate the AD4695/AD4696 (see the SPI Peripheral
Connections section).
Note that when autocycle mode is enabled, the CNV input is
ignored, and conversions are instead triggered by an internal
timer in the AD4695/AD4696, as described in the Autocycle
Mode section. When exclusively using autocycle mode, the
CNV input must be tied to IOGND. The busy indicator is
required to synchronize the AD4695/AD4696 to the SPI master
peripheral when using autocycle mode (see the SPI Peripheral
Synchronization in Autocycle Mode section).
In conversion mode, the SPI transfers must begin after tCONVERT
has elapsed and must complete within tSCKCNV before the next
CNV rising edge (see Table 2 and in the timing diagrams in the
Conversion Mode Timing Diagrams section). To ensure the
conversion mode timing requirements are met, the digital host
SPI master peripheral must either be synchronized to the clock
source generating the CNV signal or to the busy indicator
output from the AD4695/AD4696. The SCK frequency must
also be sufficiently high to ensure all conversion mode results
are clocked out before the start of the next conversion frame
(see the Conversion Mode SPI Clock Frequency Requirements
section).
SPI Peripheral Connections
The AD4695/AD4696 offer multiple serial data output modes
that allow for one or two MISO lines to output conversion
results (see the Serial Data Output Modes section). When
single-SDO mode is selected, only the SDO pin functions as a
serial data output. When dual-SDO mode is selected, both SDO
and BSY_ALT_GP0 function as serial data outputs.
CS
SCK
SDI
CS
Figure 114 shows a simplified connection diagram and software
architecture for operating the AD4695/AD4696 with only a 4-wire
SCLK
MOSI
CS
SPI. The CNV input is driven by the
host SPI master peripheral. The configuration in Figure 114
CS
output from the digital
VIO
≥2kΩ
AD4695/AD4696
SPI MASTER
requires the
signal to be periodic with deterministic rising
SDO
MISO
edge timing to achieve the necessary jitter for the application.
The SPI frames should be synchronized to a timer peripheral,
CS
Figure 112. AD4695/AD4696 SPI Connection Diagram (Single-SDO Mode)
and the output must have a well defined duty cycle. Figure 96
shows a SPI timing diagram using the configuration in Figure 114.
Figure 112 shows a connection diagram for interfacing the
AD4695/AD4696 SPI to the digital host SPI master peripheral
when configured in single-SDO mode. It is recommended to
include a pull-up resistor (2 kꢀ minimum) to VIO on the SDO
Figure 115 shows a simplified connection diagram and software
architecture for using the digital host countdown timer peripheral
to synchronize the SPI master peripheral to the CNV signal
source. The countdown timer is configured to trigger on a CNV
Rev. 0 | Page 68 of 96
Data Sheet
AD4695/AD4696
rising edge, wait for tCONVERT to elapse, and then trigger an
interrupt service routine which calls the SPI master to perform
a transfer. The countdown timer is programmed with an
integer value (count), which specifies the number of system
clock (SYS_CLK) periods to wait before calling the SPI transfer
interrupt routine. It is recommended to implement a delay
corresponding to the maximum tCONVERT specification given in
Table 2. In practice, most digital hosts exhibit some latency
between the interrupt service routine triggers and execution,
which increases the delay between the CNV rising edge and the
start of the SPI transfer. Refer to the digital host specifications
to determine the optimal count value for the given application.
the oversampled result is ready, reducing the number of
redundant SPI transfers that would otherwise occur without
additional logic (see Figure 75).
Figure 117 shows a simplified connection diagram and software
architecture for utilizing the AD4695/AD4696 threshold detection
alert indicator to synchronize the SPI master peripheral to the
ADC conversion timing. The alert indicator must be enabled
on the BSY_ALT_GP0 pin as described in the Alert Indicator
on BSY_ALT_GP0 section. The configuration in Figure 117 is
ideal in autonomous conversion applications, where the SPI is
idle while the ADC continuously converts until a user defined
out of bounds condition occurs. The alert indicator is updated
at the end of the conversion phase of the ADC and can
therefore be used as the trigger to start the SPI frame, if the SPI
frame can be completed before the start of the next conversion.
Typically, the interrupt service routine called by the alert
indicator rising edge calls the SPI to read back the conversion
result and sends the register configuration mode command
over SDI to put the AD4695/AD4696 into register
Figure 116 shows a simplified connection diagram and software
architecture for utilizing the AD4695/AD4696 busy indicator to
synchronize the SPI master peripheral to the ADC conversion
timing. The busy indicator must be enabled on the BSY_ALT_GP0
pin as described in the Busy Indicator on BSY_ALT_GP0 section,
and the digital host must have a digital input that can be
configured as a trigger for interrupt service routines. Route the
busy indicator to the interrupt input on the digital host and
configure the interrupts to trigger on the busy indicator falling
edge. Because the busy indicator falling edge is interpreted as
the data ready signal, the digital host is not required to
implement any further delays between the busy indicator
falling edge and the start of the SPI frame.
configuration mode.
The configuration in Figure 117 is ideal when operating the
AD4695/AD4696 in autocycle mode because it allows the
digital host to be completely idle until an out of bounds
condition occurs, and guarantees the digital host can remain
synchronized to the internal conversion timing (see the SPI
Peripheral Synchronization in Autocycle Mode section).
The configuration in Figure 116 is optimal when utilizing
oversampling, because the busy indicator does not go low until
tSYS_CLK
SYS_CLK
tCYC
TIMER
PERIPHERAL
PERIOD =
tSYS_CLK
CNV
CS
CS
SCK
SCLK
MOSI
MISO
AD4695/AD4696
SPI MASTER
SDI
SDO
Figure 114. 4-Wire SPI Operation Diagram
Rev. 0 | Page 69 of 96
AD4695/AD4696
Data Sheet
tSYS_CLK
SYS_CLK
CNV SOURCE
COUNTDOWN
TIMER
PERIPHERAL
tCONVERT_max
CNV
CS
COUNT =
tSYS_CLK
AD4695/AD4696 SCK
SDI
IRQ
SDOx
CS
SCLK
MOSI
MISOx
SPI MASTER
tCYC
CNV
tCONVERT
tSYS_CLK × COUNT
CS
SCK
Figure 115. SPI Synchronization with Countdown Timer Peripheral
Rev. 0 | Page 70 of 96
Data Sheet
AD4695/AD4696
CNV
BSY_ALT_GP0
CS
CNV SOURCE
AD4695/AD4696
SCK
SDI
IRQ
SDO
CS
SCLK
MOSI
MISO
SPI MASTER
tCYC
CNV
tCONVERT
BUSY
IRQ
CS
SCK
Figure 116. SPI Synchronization with Busy Indicator
CNV
CNV SOURCE
ALERT
BSY_ALT_GP0
CS
AD4695/AD4696
SCK
SDI
IRQ
SDO
CS
SCLK
MOSI
MISO
SPI MASTER
tCYC
CNV
tCNVALT
ALERT
IRQ
CS
SCK
Figure 117. SPI Synchronization with Alert Indicator
Rev. 0 | Page 71 of 96
AD4695/AD4696
Data Sheet
SPI Peripheral Synchronization in Autocycle Mode
Conversion Mode SPI Clock Frequency Requirements
If autocycle mode is enabled when the AD4695/AD4696 enter
conversion mode, the convert start instructions for the ADC
core are generated by an internal oscillator (see the Autocycle
Mode section). Autocycle mode is therefore ideal for
autonomous conversion applications, where the digital host is
idle or in a sleep state until a user programmed threshold
detection event occurs as described in the Threshold Detection
and Alert Indicators section.
Conversion results for a given sample are available until the
start of the next conversion phase. The SCK frequency must
therefore be fast enough to read the data from the AD4695/
AD4696 SPI before the following CNV rising edge (or internal
convert start signal when autocycle mode is enabled).
The minimum required SCK frequency is a function of the
sample rate in use, the length of the SPI frame (in bits), and the
serial data output mode in use. Faster sample rates require
faster SCK frequencies because the time between conversions is
shorter. Dual-SDO mode significantly reduces the required
SCK frequency for a given sample rate by doubling the number
of bits output on the SPI per SCK period (see the Serial Data
Output Modes section).
The digital host SPI must not attempt to read/write data while
the AD4695/AD4696 are still in the conversion phase. In
autocycle mode, the convert start signal is generated internally,
and the digital host must therefore reference either the busy
indicator or the alert indicator via the BSY_ALT_GP0 pin to
synchronize the AD4695/AD4696 and digital host SPIs and
ensure SPI frames occur between ADC conversion phases.
Figure 98 shows the required SPI frame timing relative to the
busy indicator and alert indicator when autocycle mode is enabled.
The number of SCK periods required per conversion mode
frame (NSCK) is a function of the number of bits per frame
(NBITS) and the number of serial data outputs (NSDO):
NSCK = NBITS/NSDO
The busy indicator can be used to trigger an interrupt service
routine to read the most recent conversion result and send the
5-bit SDI commands (see Figure 116). The busy indicator
transitions low at the end of each conversion phase and transitions
high at the start of each next conversion phase. The digital host
must begin the SPI frame following the busy indicator falling
edge, and the SCK rate must be sufficiently fast to complete the
SPI frame at least 80 ns prior to the next busy indicator rising
edge to conform to the tSCKCNV specification in Figure 98 and
Table 2 (see the Conversion Mode SPI Clock Frequency
Requirements section). The time duration between busy
indicator falling edge and rising edge is given by the tACBSY
specification in Table 2.
NBITS depends on the maximum OSR in use and whether the
status bits are enabled (see Table 19 and Table 20). NSDO is 1 for
single-SDO mode and 2 for dual-SDO mode.
The Conversion Mode Timing Diagrams section shows timing
diagrams for the SPI frames in conversion mode. The start of
the conversion mode SPI frame must not occur before the
tCONVERT
time has elapsed and must complete early enough to
adhere to the minimum tSCKCNV specification (see Table 2). The
amount of time given to complete an SPI frame in conversion
mode (tFRAME) is calculated as follows:
t
FRAME = tCYC – tCONVERT_max – tSCKCNV
1/fCNV – tCONVERT_max − tSCKCNV
where
=
The alert indicator can be used as a one-shot trigger for an
interrupt service routine on the digital host to instruct the SPI
master peripheral to send the register configuration mode
command and poll the alert registers (see Figure 117). The alert
indicator state is updated following the completion of the
conversion phase. Therefore, an alert indicator rising edge can
signify to the digital host that the AD4695/AD4696 SPI is ready
for an SPI frame. The alert indicator only transitions when a
threshold violation is detected on a given channel, however,
and therefore the digital host is not able to read conversion
results except for those that cause the alert indicator to go high
(see the Alert Indicator on BSY_ALT_GP0 section).
t
t
t
CYC is the sample period.
CONVERT_max is the maximum tCONVERT specification.
SCKCNV is the SCK to CNV rising edge delay specification (see
Table 2).
The fSCK is a function of tFRAME and NSCK
.
NBITS
f
SCK > NSCK/tFRAME =
NSDO tCYC tCONVERT tSCKCNV
Table 25 shows examples of the minimum SCK frequency
required for several sample rates for each serial data output mode
with status bits disabled and enabled and the OSR set to 1.
As described in the SPI Peripheral Synchronization in
Conversion Mode section, the digital host must complete the
SPI frame before the start of the next conversion. Refer to
Figure 98 and the Conversion Mode SPI Clock Frequency
Requirements section for guidelines on minimum SCK
frequency and overall system latency to achieve appropriate SPI
transfer rates for the selected sample rate.
When single-cycle command mode is enabled, the multiplexer
does not update channels until the 5-bit channel command is
clocked in on SDI. The SCK frequency therefore impacts tACQ
when single-cycle command mode is enabled (see the Single-
Cycle Command Mode section).
When autocycle mode is enabled, tCYC is determined by the
internal convert-start signal, the period of which is set by the
AC_CYC field, and the digital host must use either the busy
indicator or the alert indicator to synchronize the SPI frames
Rev. 0 | Page 72 of 96
Data Sheet
AD4695/AD4696
with the internal conversion timing (see the SPI Peripheral
Synchronization in Autocycle Mode section).
Note that the minimum SCK period is longer for register
configuration mode than for conversion mode (see tSCK in Table 2).
In conversion mode, the minimum tSCK is 12.5 ns, corresponding
to a maximum fSCK of 80 MHz. In register configuration mode,
the minimum tSCK is 40 ns, corresponding to a maximum fSCK of
25 MHz. Therefore, for applications requiring conversion mode
SCK frequencies of greater than 25 MHz, ensure the SPI master
peripheral serial clock rate is programmed accordingly while
the AD4695/AD4696 are in register configuration mode.
The digital host SPI master peripheral may provide more SCK
periods than required per conversion mode SPI frame. The
behavior of SDO when additional SCK falling edges occur after
the LSB is clocked out depends on the SDO_STATE bit setting.
When SDO_STATE = 0, SDO maintains its state when extra SCK
falling edges occur. When SDO_STATE = 1, SDO transitions to
high impedance when extra SCK falling edges occur.
Table 25. Minimum fSCK Requirements vs. Sample Rate and Serial Data Output Modes (OSR = 1)
Sample Rate (kSPS)
Status Bits1
Disabled
Enabled
Single-SDO Mode
Dual-SDO Mode
16 MHz
24 MHz
1000 (AD4696 Only)
1000 (AD4696 Only)
32 MHz
48 MHz
500
500
100
100
Disabled
Enabled
Disabled
Enabled
11 MHz
16 MHz
2 MHz
2.6 MHz
5.5 MHz
8 MHz
1 MHz
1.3 MHz
1 In the calculations in Table 25, NBITS = 16 when status bits are disabled and NBITS = 24 when status bits are enabled.
Rev. 0 | Page 73 of 96
AD4695/AD4696
Data Sheet
If using the standard sequencer, ensure the STD_SEQ_EN bit is
set to 1, and then program the STD_SEQ_CONFIG register
and TEMP_CTRL register to select the channels for the
sequence (see Table 49 and Table 53).
RESET
Connection Recommendations
RESET
The
input allows the digital host to trigger a full device
reset with a GPIO (see the Hardware Reset section). The
RESET
input is active low and must be driven low to initiate a
hardware reset. The AD4695/AD4696 remain in the reset state
RESET
If using the advanced sequencer, update the SEQ_CTRL
register to set the STD_SEQ_EN bit to 0 and set the
NUM_SLOTS_AS field to the desired number of advanced
sequencer slots, and program the appropriate number of
AS_SLOTn registers and TEMP_CTRL register to implement
the desired channel sequence (see Table 53 and Table 60).
until the
input is driven high.
Hardware resets are not required to operate the AD4695/AD4696,
because the SPI provides a software reset option (see the
Software Reset section). For systems not utilizing hardware
RESET
reset functionality, tie the
ensure it is pulled high during device operation.
RESET
input to VIO on board to
If using either two-cycle command mode or single-cycle
command mode, update the SEQ_CTRL register to set the
STD_SEQ_EN bit to 0 but keep the NUM_SLOTS_AS field set
to 0x0. The CYC_CTRL bit must also be set to select between two-
cycle and single-cycle command modes, but because CYC_CTRL
is in the setup register, it can be configured in the same frame as
the SPI_MODE bit is set to put the device in conversion mode.
To utilize hardware resets, connect the
or equivalent digital output from the digital host. The signal
RESET
input to a GPIO
driving
must idle high. It is recommended to also include a
RESET
weak pull-up resistor to VIO on the
pulled high until the digital host output is in a defined state. The
host firmware function for performing hardware resets must pulse
input to ensure it is
After the channel sequencing mode settings are configured,
update the CONFIG_INn register settings as needed to select
the channel configuration settings, including threshold detection
alert enable setting, polarity mode, pin pairing option, analog
input high-Z mode enable setting, and OSR. When the standard
sequencer is enabled, the settings programmed into the
CONFIG_INn register bits are applied to all analog input
channels. When any other channel sequencing mode is
selected, the settings in each CONFIG_INn register are applied
to their corresponding INn channel. See Table 54 for a detailed
description of the bits in the CONFIG_INn registers.
RESET
low following the timing requirements in Figure 99.
DEVICE CONFIGURATION RECOMMENDATIONS
The following are recommendations for configuring the desired
AD4695/AD4696 features and settings via the configuration
registers described in the Register Information section.
The AD4695/AD4696 must be in register configuration mode
to access the configuration registers via the SPI. The AD4695/
AD4696 enter register configuration mode on device power-up
and following device resets. The settings in the configuration
registers must be properly programmed for the specific application
prior to entering conversion mode and performing conversions.
When enabling threshold detection for any set of channels, update
the values in the corresponding UPPER_INn and LOWER_INn
registers to implement the desired upper and lower threshold
limits (see Table 55 and Table 56). The ALERT_MODE bit
must be updated to enable or disable hysteresis, but because
ALERT_MODE is in the SETUP register, it can be configured
in the same frame as the SPI_MODE bit is set to put the device
in conversion mode. If enabling hysteresis, the HYST_INn
registers must be updated to implement the desired hysteresis
settings.
On device power-up, it is recommended to perform either a
hardware or software reset as described in the Device Reset section.
First, program the contents of the SPI_CONFIG_A,
SPI_CONFIG_B and SPI_CONFIG_C registers to the desired
settings to ensure the AD4695/AD4696 SPI protocol is
configured to be compatible with the digital host (see the
Register Configuration Mode section). The scratch pad register
(SCRATCH_PAD) allows the digital host to validate
communications with the AD4695/AD4696 by writing test
values and reading them back without affecting device settings.
If utilizing any of the general-purpose pin functions described
in the General-Purpose Pin section, update the GPIO_CTRL
and GP_MODE register contents accordingly (see Table 50 and
Table 51).
Next, when powering VDD externally, it is recommended to
disable the internal LDO by setting the LDO_EN bit in the
setup register to 0 (see the Internal LDO section). Note that
setting the SPI_MODE bit to a 1 puts the AD4695/AD4696 into
conversion mode. Ensure that SPI_MODE is set to 0 until the
remaining configuration registers are properly configured.
If using autocycle mode, update the settings in the AC_CTRL
register to enable autocycle mode and select the desired sample
rate (see Table 48). Autocycle mode is disabled by default.
Therefore, if autocycle mode is not being used, it is not
necessary to update the AC_CTRL register after a device reset.
Next, configure the channel sequencing registers for the desired
channel sequencing mode. The SEQ_CTRL register contains
the STD_SEQ_EN bit and NUM_SLOTS_AS field, which must
be configured to select the desired channel sequencing mode.
By default, the STD_SEQ_EN bit is set to 1, which selects the
standard sequencer (see Table 47).
If utilizing offset and gain correction, update the settings in the
OFFSET_INn and GAIN_INn registers accordingly. If a
calibration routine is required to determine the necessary offset
and gain correction values for each channel, update the
Rev. 0 | Page 74 of 96
Data Sheet
AD4695/AD4696
OFFSET_INn and GAIN_INn registers after putting the AD4695/
AD4696 into conversion mode to collect enough conversion data.
When the standard sequencer is enabled, each enabled channel
in the STD_SEQ_CONFIG register is sampled once per
sequence iteration. fS_INx is therefore always constant for each
enabled channel when the standard sequencer is enabled, and is
calculated as follows:
After all other necessary configuration register settings have
been updated, put the AD4695/AD4696 in conversion mode by
setting the SPI_MODE bit in the setup register to 1. Ensure that
all other bits in the setup register are set to achieve the desired
device settings (see Table 45).
f
S_INx = fCNV/(NEN × OSR)
where
Prior to updating the setup register to put the device in
conversion mode, the digital host may optionally check the
state of the SPI_ERROR bit in the status register to verify that
there were no errors in updating the configuration registers. The
host can also check the state of the CLAMP_STATUS1 and
CLAMP_STATUS2 registers to check if any of the AD4695/
AD4696 analog input channels are experiencing overvoltage
events prior to putting the device into conversion mode.
N
EN is the number of inputs included in the channel sequence
and can range from 1 (only one channel enabled) to 17 (when
all channels and the temperature sensor are enabled).
OSR is the oversampling ratio selected by the OSR_SET field in
the CONFIG_IN0 register.
In the example provided in Figure 73, with NEN = 4 and OSR =
1, fS_INx is fCNV/4. If the OSR is programmed to 4 in this example,
fS_INx is fCNV/16.
While the AD4695/AD4696 are in conversion mode, the SPI
cannot be used to update the configuration registers. If any of
the configuration registers need to be read from or updated
while the device is already in conversion mode, send the
register configuration mode command during a conversion
mode SPI frame to put the device back into register
configuration mode (see the Register Configuration Mode
Command section).
When the advanced sequencer, two-cycle command mode or
single-cycle command mode are enabled, the sequence of
analog inputs is more flexible, and the channel sequence can be
designed to implement multiple effective sample rates. This is
useful in applications with a combination of channels with low
frequency or dc signals and channels with high frequency or ac
signals. The Implementing Two Effective Channel Sample
Rates section describes how to design a channel sequence that
achieves two effective sample rates for two sets of channels.
EFFECTIVE CHANNEL SAMPLE RATE
Table 26 and Figure 118 shows an example of a sequence which
achieves three effective sample rates with four analog inputs.
The sequence in Table 26 and Figure 118 can be implemented
with the advanced sequencer and two-cycle command mode, or
single-cycle command mode.
The AD4695/AD4696 analog inputs are multiplexed to a single
ADC core, and the state of the multiplexer is updated at the end
of the conversion phase. The effective sample rate for each
channel in the channel sequence is therefore some fraction of
the sample rate of the ADC, which is set by fCNV. The effective
sample rate for a channel is defined as the frequency at which
each new conversion result is generated for that channel.
The advanced sequencer, two-cycle command mode, and
single-cycle command mode can also be utilized to perform
aperiodic conversions on analog inputs, for example, when all
channels have dc-type signals, or when the channel sequencing
involves adaptive control logic.
For an analog input to have an effective sample rate, new
results must be generated at a constant rate for the entire
channel sequence or at least for a long enough time span to
perform the necessary analysis. For example, to calculate an
FFT and perform ac analysis on the ADC data for a given
channel, the sampling interval between each sample gathered
for that channel must be constant. The effective sample rate for
an analog input (fS_INx) is a function of fCNV and the number of
CNV periods between each time it is sampled (NCNV). The
following relationship applies for each of the 16 analog inputs
(IN0 to IN15) and for the temperature sensor as follows:
Table 26. Multiple Effective Sample Rates Example
Sequence
Position
Input
IN0
IN1
IN0
IN2
IN0
IN1
IN0
IN3
Effective Sample Rate of Input
0
1
2
3
4
5
6
7
fCNV/2
fCNV/4
fCNV/2
fCNV/8
fCNV/2
fCNV/4
fCNV/2
fCNV/8
fS_INx = fCNV/NCNV
The required fS_INx for each analog input is determined by its
input signal frequency range. The Nyquist frequency for a given
analog input (which is half of fS_INx) must be greater than the
highest signal frequency being measured to avoid aliasing.
Rev. 0 | Page 75 of 96
AD4695/AD4696
Data Sheet
...
...
CNV
IN0
IN1
IN0
IN2
IN0
IN1
IN0
IN3
IN0
IN1
IN0
IN2
IN0
IN1
IN0
IN3
IN1
SAMPLE
PERIOD
IN2
SAMPLE
PERIOD
IN3
SAMPLE
PERIOD
IN0
SAMPLE
PERIOD
Figure 118. Multiple Effective Sample Rates Example
sample rates. The number of sequence positions required (NS)
follows the relation:
Implementing Two Effective Channel Sample Rates
In multichannel data acquisition systems, the ADC may be
monitoring a mix of higher frequency and lower frequency or dc
type signals. Channels with higher maximum input frequencies
require higher Nyquist frequencies, and therefore require higher
effective sample rates than channels with lower maximum
input frequencies. To maximize the effective sample rate for
analog input channels with higher frequency input signals, the
channel sequence can be designed to implement two different
effective sample rates.
NS = NLSR × (NHSR + 1)
where
NHSR is the number of HSR inputs.
NLSR is the number of LSR inputs.
When the advanced sequencer is enabled, the maximum value
of NS is limited by the number of AS_SLOTn registers. When
two-cycle command mode or single-cycle command mode are
enabled, NS can be arbitrarily large.
In a custom channel sequence that implements two effective
sample rates, each of the AD4695/AD4696 channels included in
the sequence are categorized as either high sample rate (HSR)
channels or low sample rate (LSR) channels. Figure 119 shows a
generalized channel sequence implementing HSR and LSR
channels.
Because the LSR channels are only sampled once per full
sequence iteration, their effective sample rate (fS_LSR) is the
sample rate of the ADC core (set by fCNV) divided by NS as
follows:
f
S_LSR = fCNV/NS
N
+ 1
HSR
Because the HSR inputs are sampled once for each LSR input in
the sequence, the effective sample rate for the HSR inputs (fS_HSR) is
as follows:
HSR1
HSR1
HSR2
HSR2
HSRm
HSRm
LSR1
LSR2
f
S_HSR = (fCNV × NLSR)/NS
Table 27 shows an example where IN5, IN9, and IN14 are HSR
channels and IN2, IN10, and the temperature sensor are LSR
channels.
N
LSR
Table 27. Sequence with Two Effective Channel Sample Rates
HSR1
HSR2
HSRm
LSRn
Sequence
Position
Input
IN5
IN9
IN14
IN2
IN5
Effective Sample Rate of Input
m = N
HSR
LSR
0
1
2
3
4
5
6
7
8
9
10
11
fCNV/4
fCNV/4
fCNV/4
fCNV/12
fCNV/4
fCNV/4
fCNV/4
fCNV/12
fCNV/4
fCNV/4
fCNV/4
n = N
Figure 119. Sequence of HSR and LSR Inputs with Two Effective Sample Rates
The full channel sequence in Figure 119 consists of a repeating
sub sequence of all HSR channels, followed by one LSR channel.
The sub sequences repeat until all LSR channels are sampled
once, and then the entire sequence starts again. As a result, the
LSR channels are sampled only once per sequence iteration,
whereas the HSR channels are sampled once for each LSR
channel in the sequence.
IN9
IN14
IN10
IN5
IN9
IN14
The number of HSR channels (NHSR) and the number of LSR
channels (NLSR) dictate their effective sample rates, as well as the
number of sequence positions required to implement the two
Temperature
sensor
fCNV/12
Rev. 0 | Page 76 of 96
Data Sheet
AD4695/AD4696
Note that implementing the sequence in Table 27 with the
advanced sequencer requires the following register
configuration settings:
steps at the analog inputs. Place these external capacitors as
close to the analog inputs as possible to minimize parasitic
impedances between the two which could degrade performance.
See the Analog Front-End Design section for more
information.
STD_SEQ_EN = 0
NUM_SLOTS_AS = 10
TEMP_EN = 1
The AD4695/AD4696 voltage reference input, REF, also has a
dynamic input impedance. The effective impedance between
the reference drive circuitry output and the REF input must be
very low, and a decoupling capacitor must be placed as close to
the REF pin as possible. Connect the external reference
circuitry to the REF pin with wide traces to minimize the trace
impedance (see the Reference Circuitry Design section).
The first 11 advanced sequencer slots (AS_SLOT0 to AS_SLOT10)
are also programmed with the analog inputs listed in Table 27,
because the temperature sensor is enabled via the TEMP_EN
bit instead of via the advanced sequencer slots.
Note that when using the advanced sequencer, the temperature
sensor cannot be assigned as an HSR channel because it cannot
be assigned with the AS_SLOTn registers. However, the
temperature sensor can be included as an LSR channel by
enabling it via the TEMP_EN bit in the TEMP_CTRL register,
as demonstrated in Table 27.
The power supplies of the AD4695/AD4696 must be decoupled
with low ESR ceramic capacitors placed close to the supply
pins, and connected using short, wide traces to provide low
impedance paths and to reduce the effect of glitches on the
power supply lines (see the Power Supplies section). If LDO_IN
is powered from the same supply as AVDD, Short the pins with
a wide common trace, and a single 100 nF capacitor can be
used to decouple both pins.
LAYOUT GUIDELINES
The following are suggested layout techniques for achieving
optimal performance of the AD4695/AD4696 populated on a
printed circuit board (PCB). An example PCB layout with the
AD4696 is provided in the user guide for the AD4696
evaluation board (EVAL-AD4696FMCZ).
EVALUATING AD4695/AD4696 PERFORMANCE
The AD4695/AD4696 evaluation tool offerings include a fully
assembled and tested evaluation board including the AD4696
(EVAL-AD4696FMCZ), evaluation software for controlling the
board from a PC, and support documentation for the hardware
and software. The evaluation software requires the EVAL-SDP-
CH1Z controller board to establish communication between
the PC and the EVAL-AD4696FMCZ board.
Analog traces (that is, traces connected to the analog inputs and
reference input) must be physically separated from the digital
traces (that is, traces to the CNV input, SPI, and general-
purpose pins) to limit cross coupling from fast switching digital
signals into the analog input signals. Add ground fill between
analog and digital traces on the same PCB layer. Do not cross
digital traces over the analog traces or the AD4695/AD4696
device without a ground plane PCB layer in between. The
analog and digital pins on the AD4695/AD4696 are arranged to
facilitate separation of analog and digital traces.
The EVAL-AD4696FMCZ board allows for prototyping the
analog front-end circuitry and reference circuitry with the
various digital features offered by the AD4696. It also features a
standard 160-pin field-programmable gate array (FPGA)
mezzanine card (FMC) connector and 12-pin extended SPI
peripheral module (PMOD) connector which allow for
prototyping communication between the on-board AD4696
and many third party FPGA development boards.
The AD4695/AD4696 analog inputs (IN0 to IN15) have a
dynamic input impedance due to the multiplexer and ADC
core input switches, which toggle between conversions. An
external capacitor is recommended to reduce nonlinear voltage
Rev. 0 | Page 77 of 96
AD4695/AD4696
Data Sheet
REGISTER INFORMATION
in a single SPI transaction, or if each individual byte must be
read or written in separate SPI transactions (see the Multibyte
Register Access section).
REGISTER OVERVIEW
The AD4695/AD4696 have programmable configuration registers
that contain the bits and fields used to monitor device status and
configure the device. Reading or writing to these bits and fields
requires reading or writing to the registers that contain them. The
AD4695/AD4696 SPI is used to read and write to the configuration
registers (see the Register Configuration Mode section).
Bits and fields in the AD4695/AD4696 configuration registers
are defined as read only, read/write or R/W1C. Read only bits
can only be read from and cannot be updated by SPI writes from
the SPI master. Read/write bits can be read from or written to.
Write 1 to clear bits can be read from and are only reset to 0 when
the digital host writes a 1 in their memory location.
The AD4695/AD4696 register map memory space is divided into
bytes. Each byte of memory has a unique address, ranging from
0x000 to 0x17F. Table 28 shows the register memory address
assignments for all of the AD4695/AD4696 configuration registers.
In the access column of Table 28, registers which contain
exclusively read-only bits are represented with R and registers
with writeable bits are represented with R/W. In the access
column of Table 29 through Table 60, read only bits are
represented with R, read/write bits are represented with R/W,
and write 1 to clear bits are represented with R/W1C.
Each configuration register is a single byte or multiple bytes in
length. Registers that are multiple bytes long are called multibyte
registers. The address of each multibyte register is defined as the
address of its least significant byte, but each byte in a multibyte
register has a unique address in the register map memory space.
For example, the STD_SEQ_CONFIG register is two bytes long,
and its least significant byte (LSByte) address is 0x24 and its most
significant byte (MSByte) address is 0x25. The state of the
MB_STRICT bit in the SPI_CONFIG_C register determines
whether all bytes in a multibyte register must be read or written
The SPI_STATUS register contains various error flags that
indicate whether a SPI read or write transaction violated one of
several aspects of the protocols outlined in the Register
Configuration Mode section (see Table 37). The SPI_ERROR
bit in the status register is the bitwise logical OR of the error
flags in the SPI_STATUS register (see Table 38).
Table 28. Configuration Register Names and Descriptions
Address
0x000
0x001
0x003
0x00A
0x00C
0x00D
0x00E
0x010
0x011
0x014
0x015
0x016
0x017
0x018
0x01A
0x01B
0x020
0x021
0x022
0x023
0x024
0x026
0x027
0x028
0x029
0x030 to 0x03F
0x040 to 0x05E
0x060 to 0x07E
Name
Description
Length
Reset
0x10
0x00
0x07
0x00
0x56
0x04
0x00
0x23
0x00
0x20
0x00
0x00
0x00
0x00
0x00
0x00
0x10
0x12
0x80
0x00
0x0001
0x00
0x00
0x00
0x00
0x08
0x07FF
0x0000
Access
R/W
R/W
R
R/W
R
SPI_CONFIG_A
SPI_CONFIG_B
DEVICE_TYPE
SCRATCH_PAD
VENDOR_L
Interface Configuration A.
Interface Configuration B.
Device type.
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Single byte
Multibyte
Scratch pad.
Vendor ID (lower byte).
Vendor ID (upper byte).
Loop mode.
Interface Configuration C.
Interface status.
VENDOR_H
R
LOOP_MODE
SPI_CONFIG_C
SPI_STATUS
STATUS
ALERT_STATUS1
ALERT_STATUS2
ALERT_STATUS3
ALERT_STATUS4
CLAMP_STATUS1
CLAMP_STATUS2
SETUP
REF_CTRL
SEQ_CTRL
AC_CTRL
STD_SEQ_CONFIG
GPIO_CTRL
R/W
R/W
R/W
R
R
R
R
R
R
R
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
Device status.
Alert status (IN0 to IN3).
Alert status (IN4 to IN7).
Alert status (IN8 to IN11).
Alert status (IN12 to IN15).
Clamp status (IN0 to IN7).
Clamp status (IN8 to IN15).
Device setup.
Reference control.
Sequencer control.
Autocycle control.
Standard sequencer configuration.
GPIO enable.
General-purpose pin function control.
GPIO state.
Temperature sensor control.
Analog input settings configuration.
Upper threshold value.
Lower threshold value.
Single byte
Single byte
Single byte
Single byte
Single byte
Multibyte
GP_MODE
GPIO_STATE
TEMP_CTRL
CONFIG_INn
UPPER_INn
LOWER_INn
Multibyte
Rev. 0 | Page 78 of 96
Data Sheet
AD4695/AD4696
Address
Name
Description
Length
Reset
Access
R/W
R/W
R/W
R/W
0x080 to 0x09E
0x0A0 to 0x0BE
0x0C0 to 0x0DE
0x100 to 0x17F
HYST_INn
OFFSET_INn
GAIN_INn
AS_SLOTn
Hysteresis setting.
INn offset correction.
INn gain correction.
Advanced sequencer slot.
Multibyte
Multibyte
Multibyte
Single byte
0x0010
0x0000
0x8000
0x00
REGISTER DETAILS
SPI Configuration A Register
Address: 0x000, Reset: 0x10, Name: SPI_CONFIG_A
7
6
5
4
3
2
1
0
0
0
0
1
0
0
0
0
[7] SW_RST_MSB (R/W)
[0] SW_RST_LSB (R/W)
Software Reset Bit (MSB)
Software Reset Bit (LSB)
[6] RESERVED
[4:1] RESERVED
[5] ADDR_DIR (R/W)
Address Direction Bit
Table 29. Bit Descriptions for SPI_CONFIG_A
Bits Bit Name
Description
Reset Access
7
SW_RST_MSB Software Reset Bit (MSB). Setting both the SW_RST_MSB bit and SW_RST_LSB bit to 1 initiates a
software reset of the device, which resets all registers except the INTERFACE_CONFIG_A register
to the default power-up state (see the Software Reset section).
0x0
R/W
6
5
RESERVED
ADDR_DIR
Reserved.
0x0
0x0
R
R/W
Address Direction Bit. This bit determines sequential addressing behavior when performing
register reads and writes on multiple bytes of data in a single data phase (see the Address
Direction Options section).
0: select descending address option.
1: select ascending address option.
Reserved.
Software Reset Bit (LSB). Setting both the SW_RST_MSB and SW_RST_LSB bits to 1 initiates a
software reset of the device, which resets all registers except the INTERFACE_CONFIG_A register
to the default power-up state (see the Software Reset section).
[4:1] RESERVED
SW_RST_LSB
0x8
0x0
R
R/W
0
SPI Configuration B Register
Address: 0x001, Reset: 0x00, Name: SPI_CONFIG_B
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7] INST_MODE (R/W)
[2:0] RESERVED
Streaming or Single Instruction Mode
Select Bit
[3] ADDR_LEN (R/W)
Address Length Bit
[6:4] RESERVED
Table 30. Bit Descriptions for SPI_CONFIG_B
Bits Bit Name Description
Reset Access
7
INST_MODE Streaming or Single Instruction Mode Select Bit. This bit selects between streaming mode and
single instruction mode (see the Streaming Mode section and the Single Instruction Mode section).
0x0
R/W
0: enable streaming mode.
1: enable single instruction mode.
[6:4] RESERVED
ADDR_LEN
Reserved.
0x0
0x0
R
R/W
3
Address Length Bit. This bit sets the length of the register address in the instruction phase to
7 bits or 15 bits (see the Instruction Phase section).
0: 15-bit addressing.
1: 7-bit addressing.
Reserved.
[2:0] RESERVED
0x0
R
Rev. 0 | Page 79 of 96
AD4695/AD4696
Data Sheet
Device Type Register
Address: 0x003, Reset: 0x07, Name: DEVICE_TYPE
7
6
5
4
3
2
1
0
0
0
0
0
0
1
1
1
[7:4] RESERVED
[3:0] DEVICE_TYPE (R)
Device Type Indicator Field
Table 31. Bit Descriptions for DEVICE_TYPE
Bits Bit Name
Description
Reset Access
[7:4] RESERVED
Reserved.
0x0
0x7
R
R
[3:0] DEVICE_TYPE Device Type Indicator Field. This field identifies the Analog Devices, Inc., product category that
the device belongs to. The value 0x7 corresponds to precision ADCs.
Scratch Pad Register
Address: 0x00A, Reset: 0x00, Name: SCRATCH_PAD
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:0] SCRATCH_VALUE (R/W)
Scratchpad Field
Table 32. Bit Descriptions for SCRATCH_PAD
Bits Bit Name Description
[7:0] SCRATCH_VALUE Scratchpad Field. Values written to this register have no impact on the device behavior. Use
this register to test SPI communications with the device.
Reset Access
0x00 R/W
Vendor ID (Lower Byte) Register
Address: 0x00C, Reset: 0x56, Name: VENDOR_L
7
6
5
4
3
2
1
0
0
1
0
1
0
1
1
0
[7:0] VENDOR_ID[7:0] (R)
Vendor Identification Field
Table 33. Bit Descriptions for VENDOR_L
Bits Bit Name Description
[7:0] VENDOR_ID[7:0] Vendor Identification Field. The VENDOR_ID[15:0] field is the same value (0x0456) for all
Analog Devices precision ADCs.
Reset Access
0x56
R
Vendor ID (Upper Byte) Register
Address: 0x00D, Reset: 0x04, Name: VENDOR_H
7
6
5
4
3
2
1
0
0
0
0
0
0
1
0
0
[7:0] VENDOR_ID[15:8] (R)
Vendor Identification Field
Table 34. Bit Descriptions for VENDOR_H
Bits Bit Name Description
[7:0] VENDOR_ID[15:8] Vendor Identification Field. The VENDOR_ID[15:0] field is the same value (0x0456) for all
Analog Devices precision ADCs.
Reset Access
0x04
R
Rev. 0 | Page 80 of 96
Data Sheet
AD4695/AD4696
Loop Mode Register
Address: 0x00E, Reset: 0x00, Name: LOOP_MODE
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:0] LOOP_COUNT (R/W)
Loop Count Field
Table 35. Bit Descriptions for LOOP_MODE
Bits Bit Name Description
Reset Access
[7:0] LOOP_COUNT Loop Count Field. This field specifies the number of registers to loop through for each SPI
frame when streaming mode is selected (see the Streaming Mode section). A value of 0x00
disables looping. Values between 0x01 and 0xFF set the number of registers to loop through
before returning to the original register address.
0x00
R/W
SPI Configuration C Register
Address: 0x010, Reset: 0x23, Name: SPI_CONFIG_C
7
6
5
4
3
2
1
0
0
0
1
0
0
0
1
1
[7:6] CRC_EN (R/W)
[1:0] CRC_EN_N (R/W)
CRC Enable Field
Inverted CRC Enable Field
[5] MB_STRICT (R/W)
[4:2] RESERVED
Multibyte Access Control Bit
Table 36. Bit Descriptions for SPI_CONFIG_C
Bits Bit Name
Description
Reset Access
[7:6] CRC_EN
CRC Enable Field. This field enables the CRC when set to 0x1 (if CRC_EN_N is also set to 0x2). This
field disables the CRC when set to a value other than 0x1 (see the Checksum Protection section).
0: disables CRC.
0x0
0x1
0x0
R/W
1: enables CRC if CRC_EN_N = 0x2.
5
MB_STRICT Multibyte Access Control Bit. This bit sets the SPI transaction requirements for multibyte registers
(see the Multibyte Register Access section).
R/W
0: individual bytes in multibyte registers are read from or written to in individual data phases.
1: all bytes in multibyte registers are read from or written to in a single data phase.
Reserved.
Inverted CRC Enable Field. This field enables the CRC when set to 0x2 (if CRC_EN is also set to 0x1). 0x3
This field disables the CRC when set to a value other than 0x2 (see the Checksum Protection section).
[4:2] RESERVED
[1:0] CRC_EN_N
R
R/W
Interface Status Register
Address: 0x011, Reset: 0x00, Name: SPI_STATUS
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7] NOT_RDY_ERROR (R/W1C)
[0] ADDR_INVALID (R/W1C)
Interface Not Ready Error Flag
Invalid Address Error Flag
[6:5] RESERVED
[1] MB_ERROR (R/W1C)
Multibyte Register Access Error Flag
[4] SCK_ERROR (R/W)
SPI Clock Count Error Flag
[2] WRITE_INVALID (R/W1C)
Invalid Write Error Flag
[3] CRC_ERROR (R/W1C)
CRC Error Flag
Table 37. Bit Descriptions for SPI_STATUS
Bits Bit Name
Description
Reset Access
7
NOT_RDY_ERROR Interface Not Ready Error Flag. This bit is set to 1 when the digital host initiates an SPI
transaction before the AD4695/AD4696 interface is ready to respond, for example, before a
device reset is complete.
0x0
R/W1C
[6:5] RESERVED
Reserved.
0x0
R
Rev. 0 | Page 81 of 96
AD4695/AD4696
Data Sheet
Bits Bit Name
Description
Reset Access
4
SCK_ERROR
SPI Clock Count Error Flag. This bit is set to 1 when an incorrect number of serial clock
edges is received in an SPI read or write transaction, for example, if the SPI frame ends in
the middle of a data phase.
0x0
R/W
3
CRC_ERROR
CRC Error Flag. This bit is set to 1 when the AD4695/AD4696 receives a checksum that does
not match its expected value (see the Checksum Protection section). This error flag is only
active when the CRC is enabled.
0x0
R/W1C
2
1
WRITE_INVALID
MB_ERROR
Invalid Write Error Flag. This bit is set to 1 when the digital host attempts an SPI write to a
register that contains exclusively read only bits.
Multibyte Register Access Error Flag. This bit is set to 1 when an SPI transaction does not
access all bytes of a multibyte register. This error flag is only active when the MB_STRICT bit
is set to 1.
0x0
0x0
R/W1C
R/W1C
0
ADDR_INVALID
Invalid Address Error Flag. This bit is set to 1 when an SPI transaction attempts to access a
nonexistent register (a register with an address outside of the specified range of values in
Table 28).
0x0
R/W1C
Device Status Register
Address: 0x014, Reset: 0x20, Name: STATUS
7
6
5
4
3
2
1
0
0
0
1
0
0
0
0
0
[7] RESERVED
[0] RESERVED
[6] COM_CLAMP_FLAG (R)
[1] CLAMP_FLAG (R)
COM Overvoltage Clamp Flag
General Overvoltage Protection Clamp
Flag
[5] RESET_FLAG(R)
Reset Flag
[2] SPI_ERROR (R)
General Interface Error Flag
[4] RESERVED
[3] TD_ALERT (R)
Threshold Detection Alert Indicator
Table 38. Bit Descriptions for STATUS
Bits Bit Name Description
Reserved.
Reset Access
7
6
RESERVED
0x0
0x0
R
R
COM_CLAMP_FLAG COM Overvoltage Clamp Flag. This bit indicates if the COM overvoltage protection clamp is
active because of an overvoltage event. This bit is not sticky and is cleared when the COM
overvoltage protection clamp is inactive.
0: COM overvoltage protection clamp inactive.
1: COM overvoltage protection clamp active.
5
RESET_FLAG
Reset Flag. This bit indicates whether a hardware reset or software reset occurred since the last
time this bit was read (see the Device Reset section). This bit is automatically cleared when read.
0x1
R
0: no device reset occurred since this bit was last read.
1: a device reset occurred since this bit was last read.
Reserved.
Threshold Detection Alert Indicator. This bit indicates if any combination of the upper or
lower alert indicators for IN0 to IN15 is asserted. This bit is the logical OR of all HI_INn and
LO_INn bits in the ALERT_STATUS1 register to the ALERT_STATUS4 register. This bit is not sticky.
4
3
RESERVED
TD_ALERT
0x0
0x0
R
R
0: no upper or lower alert indicators asserted.
1: at least one upper or lower alert indicator asserted.
2
1
SPI_ERROR
General Interface Error Flag. This bit indicates if any of the error flags in the SPI_STATUS
register are asserted. This bit is the bitwise logical OR of all bits in the SPI_STATUS register.
0: no interface error detected.
0x0
R
R
1: one or more interface errors detected.
CLAMP_FLAG
General Overvoltage Protection Clamp Flag. This bit indicates if any IN0 to IN15 overvoltage 0x0
protection clamps were activated by an overvoltage event (if any of the INX_CLAMP_FLAG
bits are asserted). This bit is sticky and is only cleared if all INX_CLAMP_FLAG bits are
deasserted when the bit is read.
0: all IN0 to IN15 overvoltage clamps are inactive.
1: at least one IN0 to IN15 overvoltage clamps are active.
0
RESERVED
Reserved.
0x0
R
Rev. 0 | Page 82 of 96
Data Sheet
AD4695/AD4696
Alert Status (IN0 to IN3) Register
Address: 0x015, Reset: 0x00, Name: ALERT_STATUS1
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7] LO_IN3 (R)
[0] HI_IN0 (R)
IN3 Lower Alert Indicator
IN0 Upper Alert Indicator
[6] HI_IN3 (R)
[1] LO_IN0 (R)
IN3 Upper Alert Indicator
IN0 Lower Alert Indicator
[5] LO_IN2 (R)
[2] HI_IN1 (R)
IN2 Lower Alert Indicator
IN1 Upper Alert Indicator
[4] HI_IN2 (R)
[3] LO_IN1 (R)
IN2 Upper Alert Indicator
IN1 Lower Alert Indicator
Table 39. Bit Descriptions for ALERT_STATUS1
Bits Bit Name Description
Reset Access
7
6
5
4
3
2
1
0
LO_IN3
HI_IN3
LO_IN2
HI_IN2
LO_IN1
HI_IN1
LO_IN0
HI_IN0
IN3 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN3 is less than or equal to the IN3
lower threshold value. This indicator is only active if the threshold detection is enabled on IN3 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN3 conversion is within the range
set by the HYSTERESIS field in the HYST_IN3 register (see the Alert Indicator Registers section).
IN3 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN3 is greater than or equal to the
IN3 upper threshold value. This indicator is only active if the threshold detection is enabled on IN3 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN3 conversion is within the range
set by the HYSTERESIS field in the HYST_IN3 register (see the Alert Indicator Registers section).
IN2 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN2 is less than or equal to the IN2
lower threshold value. This indicator is only active if the threshold detection is enabled on IN2 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN2 conversion is within the range
set by the HYSTERESIS field in the HYST_IN2 register (see the Alert Indicator Registers section).
IN2 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN2 is greater than or equal to the
IN2 upper threshold value. This indicator is only active if the threshold detection is enabled on IN2 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN2 conversion is within the range
set by the HYSTERESIS field in the HYST_IN2 register (see the Alert Indicator Registers section).
IN1 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN1 is less than or equal to the IN1
lower threshold value. This indicator is only active if the threshold detection is enabled on IN1 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN1 conversion is within the range
set by the HYSTERESIS field in the HYST_IN1 register (see the Alert Indicator Registers section).
IN1 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN1 is greater than or equal to the
IN1 upper threshold value. This indicator is only active if the threshold detection is enabled on IN1 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN1 conversion is within the range
set by the HYSTERESIS field in the HYST_IN1 register (see the Alert Indicator Registers section).
IN0 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN0 is less than or equal to the IN0
lower threshold value. This indicator is only active if the threshold detection is enabled on IN0 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN0 conversion is within the range
set by the HYSTERESIS field in the HYST_IN0 register (see the Alert Indicator Registers section).
IN0 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN0 is greater than or equal to the
IN0 upper threshold value. This indicator is only active if the threshold detection is enabled on IN0 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN0 conversion is within the range
set by the HYSTERESIS field in the HYST_IN0 register (see the Alert Indicator Registers section).
0x0
0x0
0x0
0x0
0x0
0x0
0x0
0x0
R
R
R
R
R
R
R
R
Rev. 0 | Page 83 of 96
AD4695/AD4696
Data Sheet
Alert Status (IN4 to IN7) Register
Address: 0x016, Reset: 0x00, Name: ALERT_STATUS2
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7] LO_IN7 (R)
[0] HI_IN4 (R)
IN7 Lower Alert Indicator
IN4 Upper Alert Indicator
[6] HI_IN7 (R)
[1] LO_IN4 (R)
IN7 Upper Alert Indicator
IN4 Lower Alert Indicator
[5] LO_IN6 (R)
[2] HI_IN5 (R)
IN6 Lower Alert Indicator
IN5 Upper Alert Indicator
[4] HI_IN6 (R)
[3] LO_IN5 (R)
IN6 Upper Alert Indicator
IN5 Lower Alert Indicator
Table 40. Bit Descriptions for ALERT_STATUS2
Bits Bit Name Description
Reset Access
7
6
5
4
3
2
1
0
LO_IN7
HI_IN7
LO_IN6
HI_IN6
LO_IN5
HI_IN5
LO_IN4
HI_IN4
IN7 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN7 is less than or equal to the IN7
lower threshold value. This indicator is only active if the threshold detection is enabled on IN7 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN7 conversion is within the range
set by the HYSTERESIS field in the HYST_IN7 register (see the Alert Indicator Registers section).
IN7 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN7 is greater than or equal to the
IN7 upper threshold value. This indicator is only active if the threshold detection is enabled on IN7 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN7 conversion is within the range
set by the HYSTERESIS field in the HYST_IN7 register (see the Alert Indicator Registers section).
IN6 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN6 is less than or equal to the IN6
lower threshold value. This indicator is only active if the threshold detection is enabled on IN6 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN6 conversion is within the range
set by the HYSTERESIS field in the HYST_IN6 register (see the Alert Indicator Registers section).
IN6 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN6 is greater than or equal to the
IN6 upper threshold value. This indicator is only active if the threshold detection is enabled on IN6 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN6 conversion is within the range
set by the HYSTERESIS field in the HYST_IN6 register (see the Alert Indicator Registers section).
IN5 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN5 is less than or equal to the IN5
lower threshold value. This indicator is only active if the threshold detection is enabled on IN5 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN5 conversion is within the range
set by the HYSTERESIS field in the HYST_IN5 register (see the Alert Indicator Registers section).
IN5 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN5 is greater than or equal to the
IN5 upper threshold value. This indicator is only active if the threshold detection is enabled on IN5 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN5 conversion is within the range
set by the HYSTERESIS field in the HYST_IN5 register (see the Alert Indicator Registers section).
IN4 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN4 is less than or equal to the IN4
lower threshold value. This indicator is only active if the threshold detection is enabled on IN4 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN4 conversion is within the range
set by the HYSTERESIS field in the HYST_IN4 register (see the Alert Indicator Registers section).
IN4 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN4 is greater than or equal to the
IN4 upper threshold value. This indicator is only active if the threshold detection is enabled on IN4 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN4 conversion is within the range
set by the HYSTERESIS field in the HYST_IN4 register (see the Alert Indicator Registers section).
0x0
0x0
0x0
0x0
0x0
0x0
0x0
0x0
R
R
R
R
R
R
R
R
Rev. 0 | Page 84 of 96
Data Sheet
AD4695/AD4696
Alert Status (IN8 to IN11) Register
Address: 0x017, Reset: 0x00, Name: ALERT_STATUS3
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7] LO_IN11 (R)
[0] HI_IN8 (R)
IN11 Lower Alert Indicator
IN8 Upper Alert Indicator
[6] HI_IN11 (R)
[1] LO_IN8 (R)
IN11 Upper Alert Indicator
IN8 Lower Alert Indicator
[5] LO_IN10 (R)
[2] HI_IN9 (R)
IN10 Lower Alert Indicator
IN9 Upper Alert Indicator
[4] HI_IN10 (R)
[3] LO_IN9 (R)
IN10 Upper Alert Indicator
IN9 Lower Alert Indicator
Table 41. Bit Descriptions for ALERT_STATUS3
Bits Bit Name Description
Reset Access
7
LO_IN11
IN11 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN11 is less than or equal to
the IN11 lower threshold value. This indicator is only active if the threshold detection is enabled on
IN0 (see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the
ALERT_MODE bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN11
conversion is within the range set by the HYSTERESIS field in the HYST_IN11 register (see the Alert
Indicator Registers section).
0x0
R
6
5
HI_IN11
LO_IN10
IN11 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN11 is greater than or equal to
the IN11 upper threshold value. This indicator is only active if the threshold detection is enabled on IN11
(see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE
bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN11 conversion is within
the range set by the HYSTERESIS field in the HYST_IN11 register (see the Alert Indicator Registers section).
IN10 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN10 is less than or equal to the
IN10 lower threshold value. This indicator is only active if the threshold detection is enabled on IN0
(see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the
ALERT_MODE bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN10
conversion is within the range set by the HYSTERESIS field in the HYST_IN10 register (see the Alert
Indicator Registers section).
0x0
0x0
R
R
4
3
2
1
0
HI_IN10
LO_IN9
HI_IN9
LO_IN8
HI_IN8
IN10 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN10 is greater than or equal to
the IN10 upper threshold value. This indicator is only active if the threshold detection is enabled on IN10
(see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE
bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN10 conversion is within
the range set by the HYSTERESIS field in the HYST_IN10 register (see the Alert Indicator Registers section).
IN9 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN9 is less than or equal to the IN9
lower threshold value. This indicator is only active if the threshold detection is enabled on IN9 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN9 conversion is within the range
set by the HYSTERESIS field in the HYST_IN9 register (see the Alert Indicator Registers section).
IN9 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN9 is greater than or equal to the
IN9 upper threshold value. This indicator is only active if the threshold detection is enabled on IN9 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN9 conversion is within the range
set by the HYSTERESIS field in the HYST_IN9 register (see the Alert Indicator Registers section).
IN8 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN8 is less than or equal to the IN8
lower threshold value. This indicator is only active if the threshold detection is enabled on IN8 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN8 conversion is within the range
set by the HYSTERESIS field in the HYST_IN8 register (see the Alert Indicator Registers section).
0x0
0x0
0x0
0x0
0x0
R
R
R
R
R
IN8 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN8 is greater than or equal to the
IN8 upper threshold value. This indicator is only active if the threshold detection is enabled on IN8 (see the
Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE bit in the
setup register is set to 1, this bit also automatically clears if a subsequent IN8 conversion is within the range
set by the HYSTERESIS field in the HYST_IN8 register (see the Alert Indicator Registers section).
Rev. 0 | Page 85 of 96
AD4695/AD4696
Data Sheet
Alert Status (IN12 to IN15) Register
Address: 0x018, Reset: 0x00, Name: ALERT_STATUS4
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7] LO_IN15 (R)
[0] HI_IN12 (R)
IN15 Lower Alert Indicator
IN12 Upper Alert Indicator
[6] HI_IN15 (R)
[1] LO_IN12 (R)
IN15 Upper Alert Indicator
IN12 Lower Alert Indicator
[5] LO_IN14 (R)
[2] HI_IN13 (R)
IN14 Lower Alert Indicator
IN13 Upper Alert Indicator
[4] HI_IN14 (R)
[3] LO_IN13 (R)
IN14 Upper Alert Indicator
IN13 Lower Alert Indicator
Table 42. Bit Descriptions for ALERT_STATUS4
Bits Bit Name Description
Reset Access
7
LO_IN15
IN15 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN15 is less than or equal to
the IN15 lower threshold value. This indicator is only active if the threshold detection is enabled on
IN0 (see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the
ALERT_MODE bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN15
conversion is within the range set by the HYSTERESIS field in the HYST_IN15 register (see the Alert
Indicator Registers section).
0x0
R
6
5
HI_IN15
LO_IN14
IN15 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN15 is greater than or equal to
the IN15 upper threshold value. This indicator is only active if the threshold detection is enabled on IN15
(see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE
bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN15 conversion is within
the range set by the HYSTERESIS field in the HYST_IN15 register (see the Alert Indicator Registers section).
IN14 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN14 is less than or equal to
the IN14 lower threshold value. This indicator is only active if the threshold detection is enabled on
IN0 (see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the
ALERT_MODE bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN14
conversion is within the range set by the HYSTERESIS field in the HYST_IN14 register (see the Alert
Indicator Registers section).
0x0
0x0
R
R
4
3
HI_IN14
LO_IN13
IN14 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN14 is greater than or equal to
the IN14 upper threshold value. This indicator is only active if the threshold detection is enabled on IN14
(see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE
bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN14 conversion is within
the range set by the HYSTERESIS field in the HYST_IN14 register (see the Alert Indicator Registers section).
IN13 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN13 is less than or equal to
the IN13 lower threshold value. This indicator is only active if the threshold detection is enabled on
IN0 (see the Threshold Detection and Alert Indicators section). This bit is read to clear r. When the
ALERT_MODE bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN13
conversion is within the range set by the HYSTERESIS field in the HYST_IN13 register (see the Alert
Indicator Registers section).
0x0
0x0
R
R
2
1
HI_IN13
LO_IN12
IN13 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN13 is greater than or equal to
the IN13 upper threshold value. This indicator is only active if the threshold detection is enabled on IN13
(see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE
bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN13 conversion is within
the range set by the HYSTERESIS field in the HYST_IN13 register (see the Alert Indicator Registers section).
IN12 Lower Alert Indicator. This bit is set to 1 when a conversion result for IN12 is less than or equal to
the IN12 lower threshold value. This indicator is only active if the threshold detection is enabled on
IN0 (see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the
ALERT_MODE bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN12
conversion is within the range set by the HYSTERESIS field in the HYST_IN12 register (see the Alert
Indicator Registers section).
0x0
0x0
R
R
0
HI_IN12
IN12 Upper Alert Indicator. This bit is set to 1 when a conversion result for IN12 is greater than or equal to
the IN12 upper threshold value. This indicator is only active if the threshold detection is enabled on IN12
(see the Threshold Detection and Alert Indicators section). This bit is read to clear. When the ALERT_MODE
bit in the setup register is set to 1, this bit also automatically clears if a subsequent IN12 conversion is within
the range set by the HYSTERESIS field in the HYST_IN12 register (see the Alert Indicator Registers section).
0x0
R
Rev. 0 | Page 86 of 96
Data Sheet
AD4695/AD4696
Clamp Status (IN0 to IN7) Register
Address: 0x01A, Reset: 0x00, Name: CLAMP_STATUS1
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:0] INX_CLAMP_FLAG[7:0] (R)
INx Overvoltage Clamp Flags
Table 43. Bit Descriptions for CLAMP_STATUS1
Bits Bit Name Description
Reset Access
[7:0] INX_CLAMP_FLAG[7:0] INx Overvoltage Clamp Flags. This field indicates if the INx overvoltage protection
clamps are active because of an overvoltage event. Each bit corresponds to one of the
analog inputs (IN0 to IN15), where INX_CLAMP_FLAG[x] corresponds to the INx
0x0
R
overvoltage protection clamp status. INX_CLAMP_FLAG[x] is set to 1 when the INx
overvoltage protection clamp is active. These bits are not sticky and are automatically
cleared when the corresponding overvoltage protection clamp deactivates.
Clamp Status (IN8 to IN15) Register
Address: 0x01B, Reset: 0x00, Name: CLAMP_STATUS2
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:0] INX_CLAMP_FLAG[15:8] (R)
INx Overvoltage Clamp Flags
Table 44. Bit Descriptions for CLAMP_STATUS2
Bits Bit Name Description
Reset Access
0x0
[7:0] INX_CLAMP_FLAG[15:8] INx Overvoltage Clamp Flags. This field indicates if the INx overvoltage protection
clamps are active because of an overvoltage event. Each bit corresponds to one of
the analog inputs (IN0 to IN15), where INX_CLAMP_FLAG[x] corresponds to the INx
overvoltage protection clamp status. INX_CLAMP_FLAG[x] is set to 1 while the INx
overvoltage protection clamp is active. These bits are not sticky and are automatically
cleared when their corresponding overvoltage protection clamp deactivates.
R
Device Setup Register
Address: 0x020, Reset: 0x10, Name: SETUP
7
6
5
4
3
2
1
0
0
0
0
1
0
0
0
0
[7] ALERT_MODE (R/W)
[0] RESERVED
Alert Mode Select Bit
[1] CYC_CTRL (R)
[6] SDO_STATE (R)
SDO State Select Bit
Two- and Single-Cycle Command
Mode Control Bit
[5] STATUS_EN (R/W)
[2] SPI_MODE (R)
Status Bits Enable Bit
Digital Interface Mode Select Bit
[4] LDO_EN (R/W)
[3] RESERVED
Internal LDO Enable Bit
Table 45. Bit Descriptions for SETUP
Bits Bit Name Description
Reset Access
7
ALERT_MODE Alert Mode Select Bit. This bit determines how the upper and lower alert indicators (HI_INn and
LO_INn) are cleared (see the Alert Indicator Registers section).
0x0
R/W
0: hysteresis enabled.
1: hysteresis disabled.
6
SDO_STATE
SDO State Select Bit. This bit determines the behavior of serial data output(s) at the beginning
and end of conversion mode SPI frames (see the Conversion Mode Timing Diagrams section).
0x0
R/W
0: serial data output(s) hold the final value until the MSB of the next conversion data is clocked out.
1: busy indicator is enabled on the serial data output(s).
Rev. 0 | Page 87 of 96
AD4695/AD4696
Data Sheet
Bits Bit Name
Description
Reset Access
5
STATUS_EN
Status Bits Enable Bit. This bit determines whether the status bits are appended to conversion
data when in conversion mode (see the Status Bits section).
0x0
R/W
0: status bits disabled.
1: status bits enabled.
4
LDO_EN
Internal LDO Enable Bit. This bit enables or disables the internal LDO. Disable the internal LDO
when driving VDD with an external 1.8 V supply. When the internal LDO is supplying VDD,
disabling the internal LDO removes power to VDD and disables the ADC core and configuration
registers (see the Internal LDO section).
0x1
R/W
0: internal LDO disabled.
1: internal LDO enabled.
Reserved.
Digital Interface Mode Select Bit. This bit determines whether the device is in register
configuration mode or conversion mode. Set this bit to 1 to enter conversion mode. This bit is
set to 0 when the register configuration mode command is received (see the Register
Configuration Mode Command section).
3
2
RESERVED
SPI_MODE
0x0
0x0
R/W
R/W
0: selects register configuration mode.
1: selects conversion mode.
1
0
CYC_CTRL
RESERVED
Two- and Single-Cycle Command Mode Control Bit. This bit selects between two-cycle command
mode and single-cycle command mode. This bit must be set to 0 when using two-cycle command
mode, the standard sequencer, or the advanced sequencer (see the Channel Sequencing Modes
section).
0: selects two-cycle command mode.
1: selects single-cycle command mode.
Reserved.
0x0
0x0
R/W
R/W
Reference Control Register
Address: 0x021, Reset: 0x12, Name: REF_CTRL
7
6
5
4
3
2
1
0
0
0
0
1
0
0
1
0
[7] OV_MODE (R/W)
[0] RESERVED
Overvoltage Reduced Current Mode
Enable Bit
[1] REFHIZ_EN (R/W)
Reference Input High-Z Mode Enable
Bit
[6:5] RESERVED
[4:2] VREF_SET (R/W)
Reference Input Range Control
Table 46. Bit Descriptions for REF_CTRL
Bits Bit Name Description
OV_MODE
Reset Access
7
Overvoltage Reduced Current Mode Enable Bit. This bit enables or disables overvoltage reduced
current mode (see the Input Overvoltage Protection Clamps section).
0x0
R/W
0: reduce REF current during clamping.
1: do not reduce REF current during clamping.
Reserved.
Reference Input Range Control. This field configures the device to optimize performance based on 0x4
the reference voltage in use. This field must be programmed to match the VREF voltage applied to
the REF pin (see the Voltage Reference Input section).
[6:5] RESERVED
[4:2] VREF_SET
0x0
R/W
R/W
0x0: 2.4 V ≤ VREF ≤ 2.75 V.
0x1: 2.75 V < VREF ≤ 3.25 V.
0x2: 3.25 V < VREF ≤ 3.75 V.
0x3: 3.75 V < VREF ≤ 4.50 V.
0x4: 4.5 V < VREF ≤ 5.10 V.
Rev. 0 | Page 88 of 96
Data Sheet
AD4695/AD4696
Bits Bit Name
Description
Reset Access
1
REFHIZ_EN Reference Input High-Z Mode Enable Bit. This bit enables or disables reference input high-Z mode
(see the Reference Input High-Z Mode section).
0x1
R/W
0: disable reference input high-Z mode.
1: enable reference input high-Z mode.
0
RESERVED
Reserved.
0x0
R/W
Sequencer Control Register
Address: 0x022, Reset: 0x80, Name: SEQ_CTRL
7
6
5
4
3
2
1
0
1
0
0
0
0
0
0
0
[7] STD_SEQ_EN (R/W)
Standard Sequencer Enable Bit
[6:0] NUM_SLOTS_AS (R/W)
Number of Advanced Sequencer
Slots Field
Table 47. Bit Descriptions for SEQ_CTRL
Bits Bit Name Description
STD_SEQ_EN
Reset Access
7
Standard Sequencer Enable Bit. This bit enables or disables the standard sequencer (see the
Channel Sequencing Modes section).
0x1
R/W
0: standard sequencer disabled.
1: standard sequencer enabled.
[6:0] NUM_SLOTS_AS Number of Advanced Sequencer Slots Field. This field determines the number of slots in a
sequence when the advanced sequencer is enabled. The number of slots is equal to
NUM_SLOTS_AS + 1. This field must be set to 0x00 to enable two-cycle command mode or
single-cycle command mode (see the Channel Sequencing Modes section).
0x0
R/W
Autocycle Control Register
Address: 0x023, Reset: 0x00, Name: AC_CTRL
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:4] RESERVED
[3:1] AC_CYC (R/W)
[0] AC_EN (R/W)
Autocycle Mode Enable Bit
Autocycle Mode Conversion Period
Select
Table 48. Bit Descriptions for AC_CTRL
Bits Bit Name Description
[7:4] RESERVED Reserved.
Reset Access
0x0
0x0
R
R/W
[3:1] AC_CYC
Autocycle Mode Conversion Period Select. This field sets the period of the internal convert start
signal when autocycle mode is enabled (see the Autocycle Mode section).
0x0: autocycle conversion period = 10 ꢀs.
0x1: autocycle conversion period = 20 ꢀs.
0x2: autocycle conversion period = 40 ꢀs.
0x3: autocycle conversion period = 80 ꢀs.
0x4: autocycle conversion period = 100 ꢀs.
0x5: autocycle conversion period = 200 ꢀs.
0x6: autocycle conversion period = 400 ꢀs.
0x7: autocycle conversion period = 800 ꢀs.
0
AC_EN
Autocycle Mode Enable Bit. This bit enables or disables autocycle mode (see the Autocycle Mode
section).
0x0
R/W
0: autocycle mode disabled.
1: autocycle mode enabled.
Rev. 0 | Page 89 of 96
AD4695/AD4696
Data Sheet
Standard Sequencer Configuration Register
Address: 0x024, Reset: 0x0001, Name: STD_SEQ_CONFIG
15 14 13 12 11 10
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
[15] IN15_EN (R/W)
[0] IN0_EN (R/W)
IN15 Standard Sequencer Enable
Bit
IN0 Standard Sequencer Enable
Bit
[14] IN14_EN (R/W)
[1] IN1_EN (R/W)
IN14 Standard Sequencer Enable
Bit
IN1 Standard Sequencer Enable
Bit
[13] IN13_EN (R/W)
[2] IN2_EN (R/W)
IN13 Standard Sequencer Enable
Bit
IN2 Standard Sequencer Enable
Bit
[12] IN12_EN (R/W)
[3] IN3_EN (R/W)
IN12 Standard Sequencer Enable
Bit
IN3 Standard Sequencer Enable
Bit
[11] IN11_EN (R/W)
[4] IN4_EN (R/W)
IN11 Standard Sequencer Enable
Bit
IN4 Standard Sequencer Enable
Bit
[10] IN10_EN (R/W)
[5] IN5_EN (R/W)
IN10 Standard Sequencer Enable
Bit
IN5 Standard Sequencer Enable
Bit
[9] IN9_EN (R/W)
[6] IN6_EN (R/W)
IN9 Standard Sequencer Enable
Bit
IN6 Standard Sequencer Enable
Bit
[8] IN8_EN (R/W)
[7] IN7_EN (R/W)
IN8 Standard Sequencer Enable
Bit
IN7 Standard Sequencer Enable
Bit
Table 49. Bit Descriptions for STD_SEQ_CONFIG
Bits Bit Name Description
Reset Access
15
14
13
12
11
10
9
IN15_EN
IN14_EN
IN13_EN
IN12_EN
IN11_EN
IN10_EN
IN9_EN
IN8_EN
IN7_EN
IN6_EN
IN5_EN
IN4_EN
IN3_EN
IN2_EN
IN1_EN
IN0_EN
IN15 Standard Sequencer Enable Bit. When This bit is set to 1, IN15 is included in the channel
sequence when the standard sequencer is enabled (see the Standard Sequencer section).
IN14 Standard Sequencer Enable Bit. When This bit is set to 1, IN14 is included in the channel
sequence when the standard sequencer is enabled (see the Standard Sequencer section).
IN13 Standard Sequencer Enable Bit. When This bit is set to 1, IN13 is included in the channel
sequence when the standard sequencer is enabled (see the Standard Sequencer section).
IN12 Standard Sequencer Enable Bit. When This bit set to 1, IN12 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN11 Standard Sequencer Enable Bit. When This bit is set to 1, IN11 is included in the channel
sequence when the standard sequencer is enabled (see the Standard Sequencer section).
IN10 Standard Sequencer Enable Bit. When This bit is set to 1, IN10 is included in the channel
sequence when the standard sequencer is enabled (see the Standard Sequencer section).
IN9 Standard Sequencer Enable Bit. When This bit is set to 1, IN9 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN8 Standard Sequencer Enable Bit. When This bit is set to 1, IN8 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN7 Standard Sequencer Enable Bit. When This bit is set to 1, IN7 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN6 Standard Sequencer Enable Bit. When This bit is set to 1, IN6 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN5 Standard Sequencer Enable Bit. When This bit is set to 1, IN5 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN4 Standard Sequencer Enable Bit. When This bit is set to 1, IN4 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN3 Standard Sequencer Enable Bit. When This bit is set to 1, IN3 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN2 Standard Sequencer Enable Bit. When This bit is set to 1, IN2 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN1 Standard Sequencer Enable Bit. When This bit is set to 1, IN1 is included in the channel sequence 0x0
when the standard sequencer is enabled (see the Standard Sequencer section).
IN0 Standard Sequencer Enable Bit. When This bit is set to 1, IN0 is included in the channel sequence 0x1
when the standard sequencer is enabled (see the Standard Sequencer section).
0x0
0x0
0x0
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
0x0
0x0
8
7
6
5
4
3
2
1
0
Rev. 0 | Page 90 of 96
Data Sheet
AD4695/AD4696
GPIO Enable Register
Address: 0x026, Reset: 0x00, Name: GPIO_CTRL
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:5] RESERVED
[0] GPO0_EN (R/W)
BSY_ALT_GP0 GPO Enable Bit
[4] GPI0_EN (R/W)
BSY_ALT_GP0 GPI Enable Bit
[3:1] RESERVED
Table 50. Bit Descriptions for GPIO_CTRL
Bits Bit Name Description
[7:5] RESERVED Reserved.
GPI0_EN
Reset Access
0x0
0x0
R/W
R/W
4
BSY_ALT_GP0 GPI Enable Bit. This bit configures the BSY_ALT_GP0 pin as a general-purpose input
if the higher priority functions are disabled (see the General-Purpose Pin section).
0: general-purpose input function on BSY_ALT_GP0 disabled.
1: general-purpose input function on BSY_ALT_GP0 enabled.
[3:1] RESERVED Reserved.
0x0
0x0
R/W
R/W
0
GPO0_EN
BSY_ALT_GP0 GPO Enable Bit. This bit configures the BSY_ALT_GP0 pin as a general-purpose
output if all higher priority functions are disabled (see the General-Purpose Pin section).
0: general-purpose output function on BSY_ALT_GP0 disabled.
1: general-purpose output function on BSY_ALT_GP0 enabled.
General-Purpose Pin Function Control Register
Address: 0x027, Reset: 0x00, Name: GP_MODE
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7] OV_ALT_MODE (R/W)
OV_ALT Mode Select Bit
[0] ALERT_GP_EN (R/W)
Alert Indicator on General-Purpose
Pin Enable Bit
[6:4] RESERVED
[1] BUSY_GP_EN (R/W)
Busy Indicator on General-Purpose
Pin Enable Bit
[3:2] SDO_MODE (R/W)
Serial Data Output Mode Select
Table 51. Bit Descriptions for GP_MODE
Bits Bit Name Description
Reset Access
7
OV_ALT_MODE OV_ALT Mode Select Bit. This bit configures the OV_ALT bit in the status bits to report the state 0x0
of the threshold detection alert indicator (see the Status Bits section).
R/W
0: does not configure the OV_ALT bit to report the state of the TD_ALERT bit.
1: configures the OV_ALT bit to report the state of the TD_ALERT bit.
6
RESERVED
Reserved.
Reserved.
0x0
0x0
0x0
R
R/W
R/W
[5:4] RESERVED
[3:2] SDO_MODE
Serial Data Output Mode Select. This field selects the serial data output mode.
0: single-SDO mode enabled.
1: dual-SDO mode enabled.
01: single-SDO mode enabled.
11: single-SDO mode enabled.
1
0
BUSY_GP_EN
ALERT_GP_EN
Busy Indicator on General-Purpose Pin Enable Bit. This field enables or disables the busy indicator
on the BSY_ALT_GP0 pin if all higher priority functions are disabled (see the General-Purpose Pin
section).
0: busy indicator on the general-purpose pin function disabled.
1: busy indicator on the general-purpose pin function enabled.
Alert Indicator on General-Purpose Pin Enable Bit. This bit enables or disables the alert indicator on
the BSY_ALT_GP0 pin if all higher priority functions are disabled (see the General-Purpose Pin
section).
0x0
0x0
R/W
R/W
0: alert indicator on the general-purpose pin function disabled.
1: alert indicator on the general-purpose pin function enabled.
Rev. 0 | Page 91 of 96
AD4695/AD4696
Data Sheet
GPIO State Register
Address: 0x028, Reset: 0x00, Name: GPIO_STATE
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:4] GPI_READ (R)
[3:0] GPO_WRITE (R/W)
GPI State Field
GPO State Control
Table 52. Bit Descriptions for GPIO_STATE
Bits Bit Name
Description
Reset Access
[7:4] GPI_READ
GPI State Field. Bit 0 of GPI_READ displays the state of the BSY_ALT_GP0 pin when configured as a
general-purpose input.
0x0
R
[3:0] GPO_WRITE GPO State Control. Bit 0 of GPO_WRITE sets the state of the BSY_ALT_GP0 pin when configured as a
general-purpose output.
0x0
R/W
0000: BSY_ALT_GP0 is driven to a logic low voltage.
0001: BSY_ALT_GP0 is driven to a logic high voltage.
Temperature Sensor Control Register
Address: 0x029, Reset: 0x00, Name: TEMP_CTRL
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:1] RESERVED
[0] TEMP_EN (R/W)
Temperature Sensor Enable Bit
Table 53. Bit Descriptions for TEMP_CTRL
Bits Bit Name Description
Reset Access
[7:1] RESERVED Reserved.
0x0
0x0
R
0
TEMP_EN
Temperature Sensor Enable Bit. This bit enables or disables the temperature sensor in the channel
sequence when the standard sequencer or advanced sequencer is enabled (see the Temperature
Sensor section).
R/W
0: temperature sensor not included in the channel sequence.
1: temperature sensor included in the channel sequence.
Analog Input Settings Configuration Register
Address: 0x030 to Address 0x03F (Increments of 0x001), Reset: 0x08, Name: CONFIG_INn
7
6
5
4
3
2
1
0
0
0
0
0
1
0
0
0
[7] TD_EN (R/W)
[1:0] OSR_SET (R/W)
INn Threshold Detection Enable Bit
INn Oversampling Ratio Select
[6] IN_MODE (R/W)
[2] RESERVED
INn Polarity Mode Select Bit
[3] AINHIZ_EN (R/W)
[5:4] IN_PAIR (R/W)
INn Pin-Pairing Select
INn Analog Input High-Z Mode Enable
Bit
Table 54. Bit Descriptions for CONFIG_INn
Bits Bit Name Description
TD_EN
Reset Access
0x0 R/W
7
INn Threshold Detection Enable Bit. When the standard sequencer is enabled, the TD_EN bit in the
CONFIG_IN0 register enables or disables threshold detection for IN0 to IN15. When the advanced
sequencer is enabled, the TD_EN bit in each CONFIG_INn register enables or disables threshold
detection only for its corresponding INn analog input. The HI_INn and LO_INn alert indicator bits
are active when threshold detection is enabled on the corresponding INn analog input (see the
Threshold Detection and Alert Indicators section).
0: disables threshold detection for INn.
1: enables threshold detection for INn.
Rev. 0 | Page 92 of 96
Data Sheet
AD4695/AD4696
Bits Bit Name
Description
Reset Access
6
IN_MODE
INn Polarity Mode Select Bit. This bit selects the polarity mode for the corresponding INn analog
input (see the Channel Configuration Options section). Unlike the other control bits in the
CONFIG_INn registers, the polarity mode for each INn analog input is always set by the IN_MODE
bit in its corresponding CONFIG_INn register, regardless of the channel sequencing mode.
0x0
R/W
0: selects unipolar mode for INn.
1: selects pseudo bipolar mode for INn.
[5:4] IN_PAIR
INn Pin-Pairing Select. This field selects the pin pairing option for the corresponding INn analog
input (see the Channel Configuration Options section). When the standard sequencer is enabled,
the IN_PAIR field in the CONFIG_IN0 register sets the pin pairing option for IN0 to IN15. When the
advanced sequencer is enabled, the IN_PAIR bit in each CONFIG_INn register sets the pin pairing
option only for its corresponding INn analog input.
0x0
R/W
0x0: INn paired with REFGND.
0x1: INn paired with COM.
0x2: even and odd input paired.
0x3: invalid.
3
2
AINHIZ_EN INn Analog Input High-Z Mode Enable Bit. When the standard sequencer is enabled, the
AINHIZ_EN bit in the CONFIG_IN0 register enables or disables analog input high-Z mode for IN0 to
IN15. When the advanced sequencer is enabled, the AINHIZ_EN bit in each CONFIG_INn register
enables or disables analog input high-Z mode only for its corresponding INn analog input (see the
Analog Input High-Z Mode section).
0x1
R/W
0: disables analog input high-Z mode for INn.
1: enables analog input high-Z mode for INn.
RESERVED
Reserved.
0x0
0x0
R
R/W
[1:0] OSR_SET
INn Oversampling Ratio Select. When the standard sequencer is enabled, the OSR_SET field in the
CONFIG_IN0 register sets the OSR for IN0 thru IN15. When the advanced sequencer is enabled, the
OSR_SET field in each CONFIG_INn register sets the OSR only for its corresponding INn analog
input. Set the OSR_SET fields in all CONFIG_INn registers to 0x0 when two-cycle command mode
or single-cycle command mode are enabled (see the Oversampling and Decimation section).
0x0: OSR = 1 (no oversampling).
0x1: OSR = 4. Output code result resolution increases to 17 bits.
0x2: OSR = 16. Output code result resolution increases to 18 bits.
0x3: OSR = 64. Output code result resolution increases to 19 bits.
Upper Threshold Value Register
Address: 0x040 to Address 0x05E (Increments of 0x002), Reset: 0x07FF, Name: UPPER_INn
15 14 13 12 11 10
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
[15:12] RESERVED
[11:0] UPPER (R/W)
INn Upper Threshold Value Setting
Table 55. Bit Descriptions for UPPER_INn
Bits Bit Name Description
[15:12] RESERVED Reserved.
[11:0] UPPER
Reset Access
0x0
0x7FF R/W
R
INn Upper Threshold Value Setting. This field determines the upper threshold value for the
corresponding INn analog input (see the Threshold Detection and Alert Indicators section). The
value in the UPPER field corresponds to the 12 MSBs of the ADC result.
Rev. 0 | Page 93 of 96
AD4695/AD4696
Data Sheet
Lower Threshold Value Register
Address: 0x060 to Address 0x07E (Increments of 0x002), Reset: 0x0000, Name: LOWER_INn
15 14 13 12 11 10
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
[15:12] RESERVED
[11:0] LOWER (R/W)
INn Lower Threshold Value Setting
Table 56. Bit Descriptions for LOWER_INn
Bits Bit Name Description
[15:12] RESERVED Reserved.
[11:0] LOWER
Reset Access
0x0
0x0
R
R/W
INn Lower Threshold Value Setting. This field determines the lower threshold value for the
corresponding INn analog input (see the Threshold Detection and Alert Indicators section). The
value in the LOWER field corresponds to the 12 MSBs in the ADC result.
Hysteresis Setting Register
Address: 0x080 to Address 0x09E (Increments of 0x002), Reset: 0x0010, Name: HYST_INn
15 14 13 12 11 10
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
[15:12] RESERVED
[11:0] HYSTERESIS (R/W)
INn Hysteresis Value Setting
Table 57. Bit Descriptions for HYST_INn
Bits
[15:12] RESERVED
[11:0]
Bit Name
Description
Reserved.
Reset Access
0x0
R
HYSTERESIS INn Hysteresis Value Setting. This field determines the hysteresis value for the corresponding
INn analog input (see the Threshold Detection and Alert Indicators section). The value in the
HYSTERESIS field corresponds to the 12 MSBs in the ADC result.
0x10
R/W
INn Offset Correction Register
Address: 0x0A0 to Address 0x0BE (Increments of 0x002), Reset: 0x0000, Name: OFFSET_INn
15 14 13 12 11 10
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
[15:0] OFFSET (R/W)
Offset Correction Value for INn
Table 58. Bit Descriptions for OFFSET_INn
Bits
Bit Name Description
Reset Access
[15:0] OFFSET
Offset Correction Value for INn. This register sets the offset correction applied to results from the
INn channel. See the Offset and Gain Correction section for a detailed description of offset correction.
0x0
R/W
INn Gain Correction Register
Address: 0x0C0 to Address 0x0DE (Increments of 0x002), Reset: 0x8000, Name: GAIN_INn
15 14 13 12 11 10
9
8
7
6
5
4
3
2
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
[15:0] GAIN (R/W)
Gain Correction Value for INn
Table 59. Bit Descriptions for GAIN_INn
Bits
Bit Name Description
Reset
Access
[15:0] GAIN
Gain Correction Value for INn. This register sets the gain correction applied to results from the
INn channel. See the Offset and Gain Correction section for a detailed description of gain
correction.
0x8000 R/W
Rev. 0 | Page 94 of 96
Data Sheet
AD4695/AD4696
Advanced Sequencer Slot Register
Address: 0x100 to Address 0x17F (Increments of 0x001), Reset: 0x00, Name: AS_SLOTn
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
[7:4] RESERVED
[3:0] SLOT_INX (R/W)
Advanced Sequencer Slot Channel
Assignment
Table 60. Bit Descriptions for AS_SLOTn
Bits Bit Name Description
Reset Access
[7:4] RESERVED Reserved.
[3:0] SLOT_INX
0x0
0x0
R
R/W
Advanced Sequencer Slot Channel Assignment. This field determines which of the 16 analog
inputs (INx) is assigned to slot n (see the Advanced Sequencer section).
0x0: IN0.
0x1: IN1.
0x2: IN2.
0x3: IN3.
0x4: IN4.
0x5: IN5.
0x6: IN6.
0x7: IN7.
0x8: IN8.
0x9: IN9.
0xA: IN10.
0xB: IN11.
0xC: IN12.
0xD: IN13.
0xE: IN14.
0xF: IN15.
Rev. 0 | Page 95 of 96
AD4695/AD4696
Data Sheet
OUTLINE DIMENSIONS
DETAIL A
(JEDEC 95)
5.10
5.00 SQ
4.90
0.30
0.25
0.18
PIN 1
INDICATOR
AREA
PIN 1
25
32
NS
INDICATOR AR EA OP TIO
(SEE DETAIL A)
24
1
0.50
BSC
3.25
3.10 SQ
2.95
EXPOSED
PAD
17
8
9
16
0.50
0.40
0.30
0.20 MIN
TOP VIEW
SIDE VIEW
BOTTOM VIEW
0.80
0.75
0.70
FOR PROPER CONNECTION OF
THE EXPOSED PAD, REFER TO
THE PIN CONFIGURATION AND
FUNCTION DESCRIPTIONS
0.05 MAX
0.02 NOM
COPLANARITY
0.08
SECTION OF THIS DATA SHEET.
SEATING
PLANE
0.20 REF
COMPLIANT TO JEDEC STANDARDS MO-220-WHHD
Figure 120. 32-Lead Lead Frame Chip Scale Package [LFCSP]
5 mm × 5 mm Body and 0.75 mm Package Height
(CP-32-7)
Dimensions shown in millimeters
ORDERING GUIDE
Sample
Rate
Temperature
Range
Ordering Package
Quantity Option
Model1
Package Description
AD4695BCPZ
AD4695BCPZ-RL7
AD4696BCPZ
AD4696BCPZ-RL7
500 kSPS
500 kSPS
1 MSPS
−40°C to +125°C
−40°C to +125°C
−40°C to +125°C
−40°C to +125°C
32-Lead Lead Frame Chip Scale Package [LFCSP]
32-Lead Lead Frame Chip Scale Package [LFCSP]
32-Lead Lead Frame Chip Scale Package [LFCSP]
32-Lead Lead Frame Chip Scale Package [LFCSP]
490
1500
490
CP-32-7
CP-32-7
CP-32-7
CP-32-7
1 MSPS
1500
1 Z = RoHS Compliant Part.
©2020 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D24816-12/20(0)
Rev. 0 | Page 96 of 96
相关型号:
©2020 ICPDF网 联系我们和版权申明