PDF Data Sheet Rev. B

16-Bit, 5 MSPS PulSAR
Differential ADC
AD7961
Data Sheet
FEATURES
FUNCTIONAL BLOCK DIAGRAM
APPLICATIONS
REFIN
VDD1 VDD2
VIO
EN0
EN1
÷2
CLOCK
LOGIC
EN2
EN3
IN+
CAP
DAC
IN–
CNV+, CNV–
D+, D–
SERIAL
LVDS
SAR
AD7961
DCO+, DCO–
CLK+, CLK–
GND
Figure 1.
GENERAL DESCRIPTION
The AD7961 is a 16-bit, 5 MSPS, charge redistribution successive
approximation (SAR), analog-to-digital converter (ADC). The
SAR architecture allows unmatched performance both in noise
and in linearity. The AD7961 contains a low power, high speed,
16-bit sampling ADC, an internal conversion clock, and an
internal reference buffer. On the CNV± edge, the AD7961
samples the voltage difference between the IN+ and IN− pins.
The voltages on these pins swing in opposite phase between 0 V
and 4.096 V and between 0 V and 5 V. The reference voltage is
applied to the part externally. All conversion results are available
on a single LVDS self clocked or echoed clock serial interface.
The AD7961 is available in a 32-lead LFCSP (QFN) with
operation specified from −40°C to +85°C.
Table 1. Fast PulSAR® ADC Selection
Digital imaging systems
Digital X-rays
Computed tomography
IR cameras
MRI gradient control
High speed data acquisition
Spectroscopy
Test equipment
Input Type
PseudoDifferential,
16-Bit
True Bipolar,
16-Bit
Differential, 1
16-Bit
Differential,1
18-Bit
1
Rev. B
REF VCM
1 MSPS to
<2 MSPS
AD7653
AD7667
AD7980
AD7983
AD7671
2 MSPS to
3 MSPS
AD7985
5 MSPS
to 6 MSPS
AD7677
AD7623
AD7643
AD7982
AD7984
AD7621
AD7622
AD7641
AD7986
AD7625
AD7961
AD7960
10 MSPS
AD7626
Antiphase.
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 registered trademarks are the property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700 ©2013–2014 Analog Devices, Inc. All rights reserved.
Technical Support
www.analog.com
10888-001
Throughput: 5 MSPS
16-bit resolution with no missing codes
Excellent ac and dc performance
Dynamic range: 96 dB
SNR: 95.5 dB
THD: −116 dB
INL: ±0.2 LSB (typical), ±0.55 LSB (maximum)
DNL: ±0.14 LSB (typical), ±0.25 LSB (maximum)
True differential analog input voltage range: ±4.096 V or ±5 V
Low power dissipation
46.5 mW at 5 MSPS with external reference buffer
(echoed clock mode)
64.5 mW at 5 MSPS with internal reference buffer
(echoed clock mode)
39 mW at 5 MSPS with external reference buffer
(self clocked mode, CNV± in CMOS mode)
SAR architecture
No latency/pipeline delay
External reference options: 2.048 V buffered to 4.096 V (internal
reference buffer), 4.096 V, and 5 V
Serial LVDS interface
Self clocked mode
Echoed clock mode
LVDS or CMOS option for conversion control (CNV± signal)
Operating temperature range of −40°C to +85°C
32-lead, 5 mm × 5 mm LFCSP (QFN)
AD7961
Data Sheet
TABLE OF CONTENTS
Features .............................................................................................. 1 Circuit Information.................................................................... 14 Applications ....................................................................................... 1 Converter Information .............................................................. 14 Functional Block Diagram .............................................................. 1 Transfer Function ....................................................................... 15 General Description ......................................................................... 1 Analog Inputs ............................................................................. 15 Revision History ............................................................................... 2 Typical Applications ................................................................... 16 Specifications..................................................................................... 3 Voltage Reference Options ........................................................ 17 Timing Specifications .................................................................. 5 Power Supply............................................................................... 18 Absolute Maximum Ratings............................................................ 7 Digital Interface .............................................................................. 19 Thermal Resistance ...................................................................... 7 Conversion Control ................................................................... 19 ESD Caution .................................................................................. 7 Applications Information .............................................................. 22 Pin Configuration and Function Descriptions ............................. 8 Layout .......................................................................................... 22 Typical Performance Characteristics ............................................. 9 Evaluating AD7961 Performance............................................. 22 Terminology .................................................................................... 13 Outline Dimensions ....................................................................... 23 Theory of Operation ...................................................................... 14 Ordering Guide .......................................................................... 23 REVISION HISTORY
3/14—Rev. A to Rev. B
Changes to Table 4 ............................................................................ 7
Deleted Table 6; Renumbered Sequentially .................................. 7
Changes to Figure 19 ...................................................................... 11
11/13—Rev. 0 to Rev. A
Change to Table 1 ............................................................................. 1
Changes to Table 2 ............................................................................ 3
Change to Table 3 ............................................................................. 5
Changes to Table 4 ............................................................................ 7
Added Table 6; Renumbered Sequentially .................................... 7
Change to Figure 4 ........................................................................... 8
Changes to Figure 32 ...................................................................... 16
Change to Voltage Reference Options Section ........................... 17
8/13—Revision 0: Initial Version
Rev. B | Page 2 of 24
Data Sheet
AD7961
SPECIFICATIONS
VDD1 = 5 V; VDD2 = 1.8 V; VIO = 1.8 V; REF = 5 V or 4.096 V; all specifications TMIN to TMAX, unless otherwise noted.
Table 2.
Parameter
RESOLUTION
ANALOG INPUT
Voltage Range
Operating Input Voltage
Common-Mode Input Range 1
CMRR
Input Leakage Current
THROUGHPUT
Complete Cycle
Throughput Rate
DC ACCURACY
No Missing Codes
Integral Linearity Error
Differential Linearity Error
Transition Noise
Zero Error
Zero Error Drift1
Gain Error
Gain Error Drift1
Power Supply Sensitivity 2
AC ACCURACY
fIN = 1 kHz, −0.5 dBFS, VREF = 5 V
Dynamic Range
Signal-to-Noise Ratio
Spurious-Free Dynamic Range
Total Harmonic Distortion
Signal-to-Noise-and-Distortion Ratio
fIN = 1 kHz, −0.5 dBFS, VREF = 4.096 V
Dynamic Range
Signal-to-Noise Ratio
Spurious-Free Dynamic Range
Total Harmonic Distortion
Signal-to-Noise-and-Distortion Ratio
−3 dB Input Bandwidth 3
Oversampled Dynamic Range 4
Aperture Delay 5
Aperture Jitter5
REFERENCE BUFFER
REFIN Input Voltage Range1
REF Output Voltage Range
Line Regulation
Gain Drift1
Test Conditions/Comments
Min
16
VIN+ − VIN−
VIN+, VIN− to GND
−VREF
−0.1
VREF/2 − 0.05
fIN = 500 kHz
Acquisition phase
Typ
VREF/2
70
60
200
0
16
−0.55
−0.25
−2.5
−0.25
−8.5
−0.5
VDD1 = 5 V ± 5%
VDD2 = 1.8 V ± 5%
95
94.5
+VREF
VREF + 0.1
VREF/2 + 0.05
V
V
V
dB
nA
5
±0.2
±0.14
0.5
±0.01
±1
±0.05
±0.25
±0.5
+0.55
+0.25
+2.5
+0.25
+8.5
+0.5
ns
MSPS
Bits
LSB
LSB
LSB
LSB
ppm/°C
LSB
ppm/°C
LSB
LSB
dB
dB
dB
dB
dB
93
95
94.5
114
−112
94
28
115
1.6
1
dB
dB
dB
dB
dB
MHz
dB
ns
ps
2.042
4.086
2.048
4.096
2.054
4.106
V
V
−25
±20
±4
+25
µV
ppm/°C
94
93.5
EN2 = 0
OSR = 256, REF = 5 V
Rev. B | Page 3 of 24
Unit
Bits
96
95.5
118
−116
95
94
REF at 25°C, EN3 to EN0 = XX01 or
XX10
VDD1 = 5 V ± 5%, VDD2 = 1.8 V ± 5%
Max
AD7961
Parameter
EXTERNAL REFERENCE
Voltage Range
Current Drain
VCM PIN
VCM Output
VCM Error
Output Impedance
LVDS I/O (ANSI-644)
Data Format
Differential Output Voltage, VOD
Common-Mode Output Voltage, VOCM
Differential Input Voltage, VID
Common-Mode Input Voltage, VICM
POWER SUPPLIES
Specified Performance
VDD1
VDD2
VIO
Operating Currents8
Static—Not Converting, Internal
Reference Buffer Disabled
VDD1
VDD2
VIO
Static—Not Converting, Internal
Reference Buffer Enabled
VDD1
VDD2
VIO
Converting: Internal Reference Buffer
Disabled
VDD1
VDD2
VIO
Converting: Internal Reference Buffer
Enabled
VDD1
VDD2
VIO
Converting: Internal Reference Buffer
Disabled
VDD1
VDD2
VIO
Snooze Mode
VDD1
VDD2
VIO
Data Sheet
Test Conditions/Comments
Min
REFIN pin, EN1 to EN0 = 01
REF pin, EN1 to EN0 = 106
REF pin, EN1 to EN0 = 016
5 MSPS, REF = 4.096 V
5 MSPS, REF = 5 V
Typ
2.048
4.096
5
1.05
1.36
Max
Unit
1.11
1.43
V
V
V
mA
mA
REF/2
−0.01
RL = 100 Ω
RL = 100 Ω
5.1
+0.01
V
kΩ
Serial LVDS twos complement
245
290
454
9807
1130
1375
100
650
800
1575
mV
mV
mV
mV
4.75
1.71
1.71
5
1.8
1.8
5.25
1.89
1.89
V
V
V
8
8
5
40
70
5.3
μA
μA
mA
2.6
9
4.4
2.9
72
5.3
mA
μA
mA
2
11.4
9
2.2
13.5
10.3
mA
mA
mA
5.6
11.4
9
6
13.5
10.3
mA
mA
mA
2
11.4
4.9
2.2
13.5
5.6
mA
mA
mA
2
1
0.1
4.1
40.3
4.8
μA
μA
μA
Self clocked mode, CNV± in CMOS
mode9
Self clocked mode, CNV± in CMOS
mode9
Echoed clock mode, CNV± in LVDS
mode
Echoed clock mode, CNV± in LVDS
mode
Self clocked mode, CNV± in CMOS
mode9
Rev. B | Page 4 of 24
Data Sheet
Parameter
Power-Down
VDD1
VDD2
VIO
Power Dissipation
Static—Not Converting, Internal
Reference Buffer Disabled
Static—Not Converting, Internal
Reference Buffer Enabled
Converting: Internal Reference Buffer
Disabled
Converting: Internal Reference Buffer
Enabled
Converting: Internal Reference Buffer
Disabled
Power-Down
Energy per Conversion
TEMPERATURE RANGE
Specified Performance
AD7961
Test Conditions/Comments
EN3 to EN0 = X000
Min
Self clocked mode, CNV± in CMOS
mode9
Self clocked mode, CNV± in CMOS
mode9
Echoed clock mode, CNV± in LVDS
mode
Echoed clock mode, CNV± in LVDS
mode
Self clocked mode, CNV± in CMOS
mode9
EN3 to EN0 = X000
Self clocked, CNV± in CMOS mode9
TMIN to TMAX
Typ
Max
Unit
1
1
0.2
2.8
37.8
4.6
μA
μA
μA
9
10.3
mW
21
25
mW
46.5
56.2
mW
64.5
76.4
mW
39
47.4
mW
7.2
7.8
94.5
9.5
μW
nJ/sample
+85
°C
−40
1
The minimum and maximum values are guaranteed by characterization.
Using an external reference.
3
See Table 9 for logic levels of enable pins. When EN2 = 1, the −3 dB input bandwidth is 9 MHz. Use this lower bandwidth only when the throughput rate is 2 MSPS or
lower.
4
The oversampled dynamic range is the ratio of the peak signal power to the noise power (for a small input) measured in the ADC output FFT from dc up to fS/(2 × OSR),
where fS is the ADC sample rate and OSR is the oversampling ratio.
5
Guaranteed by design.
6
The REFIN pin is tied to 0 V in this mode.
7
The ANSI-644 LVDS specification has a minimum common-mode output (VOCM) of 1125 mV.
8
The current dissipated in the VCM circuitry when enabled is REF/20 kΩ and is not included in the operating currents listed.
9
CNV+ works as a CMOS input when CNV− is grounded. See Table 7 for additional information.
2
TIMING SPECIFICATIONS
VDD1 = 5 V; VDD2 = 1.8 V; VIO = 1.71 V to 1.89 V; REF = 5 V or 4.096 V; all specifications TMIN to TMAX, unless otherwise noted.
Table 3.
Parameter
Time Between Conversions
Acquisition Time
CNV± High Time
CNV± to D± (MSB) Ready
CNV± to Last CLK± (LSB) Delay
CLK± Period1
CLK± Frequency
CLK± to DCO± Delay (Echoed Clock Mode)
DCO± to D± Delay (Echoed Clock Mode)
CLK± to D± Delay
1
Symbol
tCYC
tACQ
tCNVH
tMSB
tCLKL
tCLK
fCLK
tDCO
tD
tCLKD
Min
200
Typ
Max
tCYC − 115
10
3.33
0
0
4
250
3
0
3
0.6 × tCYC
200
160
(tCYC − tMSB + tCLKL)/n
300
5
1
5
Unit
ns
ns
ns
ns
ns
ns
MHz
ns
ns
ns
For the maximum CLK± period, the window available to read data is tCYC − tMSB + tCLKL. Divide this time by the number of bits (n) to be read giving the maximum CLK±
frequency that can be used for a given conversion CNV± frequency. In echoed clock interface mode, n = 16; in self clocked interface mode, n = 18.
Rev. B | Page 5 of 24
AD7961
Data Sheet
Timing Diagrams
SAMPLE N
SAMPLE N + 1
tCYC
tCNVH
CNV–
CNV+
tACQ
ACQUISITION
ACQUISITION
ACQUISITION
tCLKL
tCLK
15
CLK–
16
1
2
15
16
1
2
3
CLK+
tDCO
15
DCO–
16
1
2
15
1
16
2
3
DCO+
tMSB
D+
D1
N–1
D–
D0
N–1
tD
D15
N
0
D14
N
D1
N
D0
N
0
D15
N+1
D14
N+1
10888-002
tCLKD
D13
N+1
Figure 2. Echoed Clock Interface Mode Timing Diagram
SAMPLE N
SAMPLE N + 1
tCYC
tCNVH
CNV–
CNV+
tACQ
ACQUISITION
ACQUISITION
ACQUISITION
tCLKL
tCLK
CLK–
17
18
1
2
4
3
17
1
18
2
3
CLK+
D+
D–
D1
N–1
D0
N–1
0
1
0
D15
N
D14
N
D1
N
Figure 3. Self-Clocked Interface Mode Timing Diagram
Rev. B | Page 6 of 24
D0
N
0
1
0
D15
N+1
10888-003
tMSB
tCLKD
Data Sheet
AD7961
ABSOLUTE MAXIMUM RATINGS
THERMAL RESISTANCE
Table 4.
Parameter
Analog Inputs/Outputs
IN+, IN− to GND
REF1 to GND
VCM to GND
REFIN to GND
Supply Voltages
VDD1
VDD2, VIO
Digital Inputs to GND
Digital Outputs to GND
Input Current to Any Pin
Except Supplies
Operating Temperature
Range (Commercial)
Storage Temperature Range
Junction Temperature
ESD Ratings
Human Body Model
Machine Model
Field-Induced ChargedDevice Model
1
θJA is specified for the worst-case conditions, that is, a device
soldered in a circuit board for surface-mount packages.
Rating
−0.3 V to VDD1
−0.3 V to +6 V
−0.3 V to +6 V
−0.3 V to +6 V
Table 5. Thermal Resistance
Package Type
32-Lead LFCSP_VQ
−0.3 V to +6 V
−0.3 V to +2.1 V
−0.3 V to VIO + 0.3 V
−0.3 V to VIO + 0.3 V
±10 mA
ESD CAUTION
−40°C to +85°C
−65°C to +150°C
150°C
4 kV
200 V
1.25 kV
Transient currents of up to 100 mA do not cause SCR latch-up.
Stresses above those listed under Absolute Maximum Ratings
may cause permanent damage to the device. This is a stress
rating only; functional operation of the device at these or any
other conditions above those indicated in the operational
section of this specification is not implied. Exposure to absolute
maximum rating conditions for extended periods may affect
device reliability.
Rev. B | Page 7 of 24
θJA
40
θJC
4
Unit
°C/W
AD7961
Data Sheet
32
31
30
29
28
27
26
25
REF
REF
REF
REF
REF_GND
REF_GND
REF_GND
VDD2
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
1
2
3
4
5
6
7
8
AD7961
TOP VIEW
(Not to Scale)
24
23
22
21
20
19
18
17
GND
IN+
IN–
VCM
VDD1
VDD1
VDD2
CLK+
NOTES
1. CONNECT THE EXPOSED PAD TO THE
GROUND PLANE OF THE PCB
USING MULTIPLE VIAS.
10888-004
CNV+
D–
D+
VIO
GND
DCO–
DCO+
CLK–
9
10
11
12
13
14
15
16
VDD1
VDD2
REFIN
EN0
EN1
EN2
EN3
CNV–
Figure 4. Pin Configuration
Table 7. Pin Function Descriptions
Pin No.
1, 19, 20
2, 18, 25
12
13, 24
26, 27, 28
Mnemonic
VDD1
VDD2
VIO
GND
REF_GND
Type 1
P
P
P
P
P
3
REFIN
AI
4, 5, 6, 7
DI
8, 9
EN0, EN1,
EN2, 2 EN3
CNV−, CNV+
10, 11
14, 15
D−, D+
DCO−, DCO+
DO
DO
16, 17
21
CLK−, CLK+
VCM
DI
AO
22
23
29, 30, 31,
32
IN−
IN+
REF
AI
AI
AI/O
33
EP
1
2
DI
Description
Analog 5 V Supply. Decouple the 5 V supply with a 100 nF capacitor.
Analog 1.8 V Supply. Decouple this pin with a 100 nF capacitor.
Input/Output Interface Supply. Use a 1.8 V supply and decouple this pin with a 100 nF capacitor.
Ground.
Reference Ground. Connect the capacitors on the REF pin between REF and REF_GND. Tie REF_GND to
GND.
Prebuffer Reference Voltage. It is driven with an external reference voltage of 2.048 V. When driving an
external 2.048 V reference, a 100 nF capacitor is required. If using an external 5 V or 4.096 V reference
(connected to REF), connect this pin to ground.
Enable.2 The logic levels of these pins set the operation of the device as described in Table 9.
Convert Input. These pins act as the conversion control pin. On the rising edge of these pins, the analog
inputs are sampled and a conversion cycle is initiated. CNV+ works as a CMOS input when CNV− is
grounded; otherwise, CNV+ and CNV− are differential LVDS inputs.
LVDS Data Outputs. The conversion data is output serially on these pins.
LVDS Buffered Clock Outputs. When DCO+ is grounded, the self-clocked interface mode is selected. In this
mode, the 16-bit results on D± are preceded by an initial 0 (which is output at the end of the previous
conversion), followed by a 2-bit header (10) to allow synchronization of the data by the digital host with
extra logic. The 1 in this header provides the reference to acquire the subsequent conversion result
correctly. When DCO+ is not grounded, the echoed clock interface mode is selected. In this mode, DCO± is
a copy of CLK±. The data bits are output on the falling edge of DCO+ and can be captured in the digital
host on the next rising edge of DCO+.
LVDS Clock Inputs. This clock shifts out the conversion results on the falling edge of CLK+.
Common-Mode Output. When using any reference scheme, this pin produces one-half the voltage present
on the REF pin, which can be useful for driving the common mode of the input amplifiers.
Differential Negative Analog Input. Referenced to and must be driven 180° out of phase with IN+.
Differential Positive Analog Input. Referenced to and must be driven 180° out of phase with IN−.
Buffered Reference Voltage. When using the 2.048 V external reference (REFIN input), the 4.096 V system
reference is produced at this pin. When using an external reference of 4.096 V or 5 V on this pin, the
internal reference buffer must be disabled. Connect the REF pins with the shortest trace possible to a
single 10 μF, low ESR, low ESL capacitor. The other side of the capacitor must be placed close to GND.
Exposed Pad. The exposed pad is located on the underside of the package. Connect the exposed pad to
the ground plane of the PCB using multiple vias.
AI = analog input; AI/O = bidirectional analog; AO = analog output; DI = digital input; DO = digital output; P = power.
EN2 = 0 sets the 28 MHz of input bandwidth and EN2 = 1 sets the 9 MHz of input bandwidth. EN3 = 1 enables the VCM reference output.
Rev. B | Page 8 of 24
Data Sheet
AD7961
TYPICAL PERFORMANCE CHARACTERISTICS
VDD1 = 5 V; VDD2 = 1.8 V; VIO = 1.8 V; all specifications T = 25°C, unless otherwise noted.
0.3
0.2
–40°C
+25°C
+85°C
–40°C
+25°C
+85°C
0.2
0.1
DNL (LSB)
INL (LSB)
0.1
0
0
–0.1
–0.1
–0.3
0
10000
20000
30000
40000
CODE
50000
–0.2
60000
10888-101
10888-100
–0.2
0
10000
20000
30000
40000
50000
60000
CODE
Figure 5. Integral Nonlinearity vs. Code and Temperature, REF = 5 V
Figure 8. Differential Nonlinearity vs. Code and Temperature, REF = 5 V
0.2
0.3
–40°C
+25°C
+85°C
–40°C
+25°C
+85°C
0.2
0.1
DNL (LSB)
INL (LSB)
0.1
0
0
–0.1
–0.1
–0.3
0
10000
20000
30000
40000
50000
–0.2
60000
10888-103
10888-102
–0.2
0
10000
20000
30000
40000
50000
60000
CODE
CODE
Figure 6. Integral Nonlinearity vs. Code and Temperature, REF = 4.096 V
Figure 9. Differential Nonlinearity vs. Code and Temperature, REF = 4.096 V
150000
250000
216380
128593
125000
200000
116886
100000
COUNT
COUNT
150000
75000
100000
50000
0
57
2C1
2C2
0
2C3
25000
20940
2C5
2C4
CODE (HEX)
66
0
2C6
2C7
10888-104
24701
0
16577
0
86
2C0
2C1
2C2
2C3
2C4
2
0
2C5
2C6
CODE (HEX)
Figure 7. Histogram of DC Input at Code Center, REF = 5 V
Figure 10. Histogram of DC Input at Code Transition, REF = 5 V
Rev. B | Page 9 of 24
10888-105
50000
AD7961
Data Sheet
160000
250000
215449
136440
140000
124393
200000
120000
100000
COUNT
COUNT
150000
80000
100000
60000
40000
50000
E56F
E570
E571
E572
E573
CODE (HEX)
1
0
E574
E575
20000
776
E56F
E570
0
0
E571
E572
E573
E574
E575
CODE (HEX)
Figure 14. Histogram of DC Input at Code Transition, REF = 4.096 V
0
0
INPUT FREQENCY = 20kHz
SNR = 95.9dB
SINAD = 95.8dB
THD = –115.5dB
SFDR = 117dB
–40
–20
–40
AMPLITUDE (dB)
–60
–80
–100
–120
–60
–80
–100
–120
–140
–160
–160
10888-106
–140
–180
0
1.0
0.5
1.5
2.0
INPUT FREQENCY = 20kHz
SNR = 96.2dB
SINAD = 96.1dB
THD = –121dB
SFDR = 122dB
10888-110
–20
–180
2.5
0
0.5
FREQUENCY (MHz)
1.0
1.5
2.0
2.5
FREQUENCY (MHz)
Figure 12. 20 kHz, −0.5 dBFS Input Tone FFT, Wide View, REF = 5 V
Figure 15. 20 kHz, −6 dBFS Input Tone FFT, Wide View, REF = 5 V
0
0
–40
–40
AMPLITUDE (dB)
–60
–80
–100
–120
–60
–80
–100
–120
–140
–160
–160
10888-108
–140
–180
0
10
20
30
40
50
60
70
80
90
INPUT FREQENCY = 20kHz
SNR = 95.2dB
SINAD = 95.1dB
THD = –110.8dB
SFDR = 113.4dB
–20
INPUT FREQENCY = 20kHz
SNR = 95.9dB
SINAD = 95.8dB
THD = –115.5dB
SFDR = 117dB
10888-107
–20
AMPLITUDE (dB)
535
0
Figure 11. Histogram of DC Input at Code Center, REF = 4.096 V
AMPLITUDE (dB)
0
10888-127
0
3
10888-126
24360
22331
0
–180
0
100
FREQUENCY (kHz)
Figure 13. 20 kHz, −0.5 dBFS Input Tone FFT, Zoomed View, REF = 5 V
0.5
1.0
1.5
FREQUENCY (MHz)
2.0
2.5
Figure 16. 20 kHz, −0.5 dBFS Input Tone FFT, Wide View, REF = 4.096 V
Rev. B | Page 10 of 24
Data Sheet
AD7961
96.0
0
–20
INPUT FREQENCY = 20kHz
SNR = 95.2dB
SINAD = 95.1dB
THD = –110.8dB
SFDR = 113.4dB
95.8
–60
SNR, SINAD (dB)
–80
–100
–120
–140
SNR
SINAD
95.6
95.4
95.2
10888-109
–160
–180
0
10
20
30
40
50
60
70
80
90
95.0
–40
100
10888-114
AMPLITUDE (dB)
–40
–20
0
FREQUENCY (kHz)
60
80
Figure 20. SNR and SINAD vs. Temperature, REF = 5 V
Figure 17. 20 kHz, −0.5 dBFS Input Tone FFT, Zoomed View, REF = 4.096 V
–110
0
–20
INPUT FREQENCY = 20kHz
SNR = 95.5dB
SINAD = 95.4dB
THD = –119.9dB
SFDR = 119.7dB
–40
–112
–114
–60
THD (dB)
AMPLITUDE (dB)
20
40
TEMPERATURE (°C)
–80
–100
–116
–118
–120
–140
10888-111
–180
0
0.5
1.0
1.5
2.0
–122
–40 –30 –20 –10
2.5
0
FREQUENCY (MHz)
10
20
30
40
50
60
70
80
70
80
TEMPERATURE (°C)
Figure 21. THD vs. Temperature, REF = 5 V
Figure 18. 20 kHz, −6 dBFS Input Tone FFT, Wide View, REF = 4.096 V
–120
96.00
SNR
THD
10888-128
–120
–160
126
–115
124
–110
95.75
122
SFDR (dB)
–100
–95
95.00
–80
0
50
100
150
10888-130
–85
118
116
–90
95.25
120
200
FREQUENCY (kHz)
114
112
–40 –30 –20 –10
0
10
20
30
40
50
60
TEMPERATURE (°C)
Figure 19. SNR and THD vs. Frequency, −0.5 dBFS, REF = 5 V
Figure 22. SFDR vs. Temperature, REF = 5 V
Rev. B | Page 11 of 24
10888-129
95.50
THD (dB)
SNR (dB)
–105
AD7961
Data Sheet
10
0.9
VDD2
VDD1
VIO
8
0.8
0.7
CURRENT (µA)
GAIN ERROR
0.6
0.5
0.4
6
4
0.3
0.2
0.1
ZERO ERROR
0
–40
–20
0
20
40
TEMPERATURE (̊°C)
60
0
–40
80
10888-119
2
10888-120
ZERO ERROR AND GAIN ERROR (LSB)
1.0
–20
0
20
40
60
80
TEMPERATURE (°C)
Figure 23. Zero Error and Gain Error vs. Temperature, REF = 5 V
Figure 26. Power-Down Current vs. Temperature, REF = 5 V
0.3
12
0.2
10
0
SUPPLY CURRENT (mA)
IN+
–0.1
–0.2
IN–
–0.3
–0.4
–0.5
6
VIO
4
2
VDD1
10888-122
–0.6
–0.7
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
DIFFERENTIAL INPUT VOLTAGE (V)
4.5
0
5.0
VDD2
10
8
6
VIO
4
VDD1
0
–40
10888-118
2
–20
0
20
40
60
1
2
3
4
5
Figure 27. Supply Current vs. Throughput, Self Clocked Mode, CNV± in CMOS
Mode, Internal Reference Buffer Disabled
14
12
0
THROUGHPUT (MHz)
Figure 24. Input Current (IN+, IN−) vs. Differential Input Voltage, REF = 5 V
SUPPLY CURRENT (mA)
VDD2
8
10888-124
INPUT CURRENT (mA)
0.1
80
TEMPERATURE (°C)
Figure 25. Supply Current vs. Temperature, REF = 5 V, Self Clocked Mode,
CNV± in CMOS Mode, Internal Reference Buffer Disabled
Rev. B | Page 12 of 24
Data Sheet
AD7961
TERMINOLOGY
Differential Nonlinearity (DNL) Error
In an ideal ADC, code transitions are 1 LSB apart. Differential
nonlinearity is the maximum deviation from this ideal value. It
is often specified in terms of resolution for which no missing
codes are guaranteed.
Integral Nonlinearity (INL) Error
Linearity error refers to the deviation of each individual code
from a line drawn from negative full scale through positive full
scale. The point used as negative full scale occurs ½ LSB before
the first code transition. Positive full scale is defined as a level
1½ LSB beyond the last code transition. The deviation is measured from the middle of each code to the true straight line.
Dynamic Range
Dynamic range is the ratio of the rms value of the full scale to
the rms noise measured for an input typically at −60 dB. The
value for dynamic range is expressed in decibels.
Effective Number of Bits (ENOB)
ENOB is a measurement of the resolution with a sine wave
input. It is related to SINAD and is expressed in bits by
ENOB = [(SINADdB − 1.76)/6.02]
Gain Error
The first transition (from 100 … 000 to 100 …001) should occur
at a level ½ LSB above nominal negative full scale (−4.0959844 V
for the ±4.096 V range). The last transition (from 011 … 110 to
011 … 111) occurs for an analog voltage 1½ LSB below the
nominal full scale (+4.095953 V for the ±4.096 V range). The
gain error is the deviation of the difference between the actual
level of the last transition and the actual level of the first
transition from the difference between the ideal levels.
Gain Error Drift
The ratio of the gain error change due to a temperature change
of 1°C and the full-scale range (2N). It is expressed in parts per
million.
Power Supply Rejection Ratio (PSRR)
Variations in power supply affect the full-scale transition but not
the linearity of the converter. PSRR is the maximum change in
the full-scale transition point due to a change in power supply
voltage from the nominal value.
Signal-to-Noise Ratio (SNR)
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 decibels.
Signal-to-Noise-and-Distortion (SINAD) Ratio
SINAD 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, including harmonics but excluding dc. The value for
SINAD is expressed in decibels.
Spurious-Free Dynamic Range (SFDR)
SFDR is the difference, in decibels, between the rms amplitude
of the input signal and the peak spurious signal (including
harmonics).
Total Harmonic Distortion (THD)
THD is the ratio of the rms sum of the first five harmonic
components to the rms value of a full-scale input signal and
is expressed in decibels.
Zero Error
Zero error is the difference between the ideal midscale input
voltage (0 V) and the actual voltage producing the midscale
output code.
Zero Error Drift
The ratio of the zero error change due to a temperature change
of 1°C and the full scale code range (2N). It is expressed in parts
per million.
Least Significant Bit (LSB)
The least significant bit, or LSB, is the smallest increment that
can be represented by a converter. For a fully differential input
ADC with N bits of resolution, the LSB expressed in volts is
LSB (V) =
V INp-p
2N
Rev. B | Page 13 of 24
AD7961
Data Sheet
THEORY OF OPERATION
IN+
GND
LSB
MSB
32,768C
16,384C
4C
2C
C
SWITCHES
CONTROL
SW+
C
CLK+, CLK–
REF
(4.096V)
COMP
CONTROL
LOGIC
4C
2C
C
OUTPUT CODE
C
MSB
SW–
LSB
CNV+, CNV–
GND
LVDS INTERFACE
CONVERSION
CONTROL
IN–
10888-011
16,384C
DATA
TRANSFER
D+, D–
GND
32,768C
DCO+, DCO–
Figure 28. ADC Simplified Schematic
CIRCUIT INFORMATION
The AD7961 is a 5 MSPS, high precision, power efficient, 16-bit
ADC that uses SAR-based architecture to provide performance
of 95.5 dB SNR, ±0.2 LSB INL, and ±0.14 LSB DNL. The AD7961
does not exhibit any pipeline delay or latency, making it ideal
for multiplexed channel applications.
The AD7961 is capable of converting 5,000,000 samples per
second (5 MSPS). The device typically consumes 46.5 mW of
power. The AD7961 offers the added functionality of an onchip reference buffer. If the internal reference buffer is enabled,
the AD7961 consumes approximately an additional 18 mW of
power.
The AD7961 is specified for use with 5 V and 1.8 V supplies
(VDD1, VDD2). The interface from the digital host to the
AD7961 uses 1.8 V logic only. The AD7961 uses an LVDS
interface to transfer data conversions. The CNV+ and CNV−
inputs to the part activate the conversion of the analog input.
The CNV+ and CNV− pins can be applied using a CMOS or
LVDS source.
The AD7961 is housed in a space-saving, 32-lead, 5 mm ×
5 mm LFCSP package.
When the conversion phase begins, SW+ and SW− are opened
first. The two-capacitor arrays are then disconnected from the
inputs and connected to the GND input. Therefore, the differential
voltage between the inputs (IN+ and IN−) captured at the end
of the acquisition phase is applied to the comparator inputs,
causing the comparator to become unbalanced. By switching
each element of the capacitor array between GND and REF
(the reference voltage), the comparator input varies by binary
weighted voltage steps (VREF/2, VREF/4 … VREF/262,144). The
control logic toggles these switches, MSB first, to bring the
comparator back into a balanced condition. At the completion
of this process, the control logic generates the ADC output code.
The AD7961 digital interface uses low voltage differential
signaling (LVDS) to enable high data transfer rates.
The AD7961 conversion result is available for reading after tMSB
(time from the conversion start until MSB is available) elapses.
The user must apply a burst LVDS CLK± signal to the AD7961
to transfer data to the digital host.
The CLK± signal outputs the ADC conversion result onto the
data output D±. The bursting of the CLK± signal, illustrated in
Figure 35 and Figure 36, is characterized as follows:
CONVERTER INFORMATION
•
The AD7961 is a 5 MSPS ADC that uses SAR-based architecture based on a charge redistribution DAC. Figure 28 shows
a simplified schematic of the ADC. The capacitive DAC consists
of two identical arrays of 16 binary weighted capacitors that are
connected to the two comparator inputs.
•
During the acquisition phase, the terminals of the array tied
to the input of the comparator are connected to GND via SW+
and SW−. All independent switches are connected to the analog
inputs. In this way, the capacitor arrays are used as sampling
capacitors and acquire the analog signal on the IN+ and IN−
inputs. A conversion phase is initiated when the acquisition
phase is complete and the CNV± input goes high. Note that the
AD7961 can receive a CMOS or LVDS format CNV± signal.
Rev. B | Page 14 of 24
Hold the differential voltage on CLK± in a steady state in
the window of time between tCLKL and tMSB.
The AD7961 has two data read modes. For more
information about the echoed clock and self clocked
interface modes, see the Digital Interface section.
Data Sheet
AD7961
TRANSFER FUNCTION
The AD7961 uses a 5 V or a 4.096 V reference. The AD7961
converts the differential voltage of the antiphase analog inputs
(IN+ and IN−) into a digital output. IN+ and IN− require a
REF/2 V common-mode voltage.
maximum. However, if the supplies of the input buffer amplifier
are different from the VDD1/GND supply, the analog input
signal may eventually exceed the supply rails by more than
0.3 V. In such a case (for example, an input buffer with a short
circuit), the current limitation can be used to protect the part.
VDD1
185Ω
26pF
10888-013
IN+
OR IN–
Figure 30. Equivalent Analog Input Circuit
011 ... 111
011 ... 110
011 ... 101
The analog input structure allows the sampling of the true
differential signal between IN+ and IN−. By using these differential inputs, signals common to both inputs are rejected. The
AD7961 shows some degradation in THD with higher analog
input frequencies.
100
100 ... 010
90
80
–FSR + 0.5LSB
+FSR – 1LSB
+FSR – 1.5LSB
ANALOG INPUT
70
Figure 29. ADC Ideal Transfer Functions (FSR = Full-Scale Range)
ANALOG INPUTS
The analog inputs applied to the AD7961, IN+ and IN−, must be
180° out of phase with each other. Figure 30 shows an equivalent
circuit of the input structure of the AD7961.
The two diodes provide ESD protection for IN+ and IN−. Care
must be taken to ensure that the analog input signals do not exceed
the supply rails of the AD7961 by more than 0.3 V (VDD1 and
GND). If the analog input signals exceed this level, the diodes
become forward-biased and start conducting current. These
diodes can handle a forward-biased current of 130 mA
60
50
40
30
20
10888-123
–FSR + 1LSB
CMRR (dB)
100 ... 001
100 ... 000
–FSR
10888-012
ADC CODE (TWOS COMPLEMENT)
The 16-bit conversion result is in MSB first, twos complement
format. The ideal transfer functions for the AD7961 are shown
in Figure 29 and Table 8.
10
0
100
1k
10k
100k
1M
FREQUENCY (Hz)
Figure 31. Analog Input CMRR vs. Frequency
Table 8. Output Codes and Ideal Input Voltages
Description
FSR − 1 LSB
Midscale + 1 LSB
Midscale
Midscale − 1 LSB
−FSR + 1 LSB
−FSR
Analog Input (IN+ − IN−),
REF = 5 V
+4.999847 V
+152.6 μV
0V
−152.6 μV
−4.999847 V
−5 V
Analog Input (IN+ − IN−),
REF = 4.096 V
+4.095875 V
+125 μV
0V
−125 μV
−4.095875 V
−4.096 V
Rev. B | Page 15 of 24
Digital Output Code, Twos Complement (Hex)
0x7FFF
0x0001
0x0000
0xFFFF
0x8001
0x8000
AD7961
Data Sheet
edge to the multiplexer inputs switching event results in no
corruption. If the analog inputs are multiplexed during this
quiet conversion time, the current conversion may be corrupted
by up to 4 LSBs.
TYPICAL APPLICATIONS
Figure 32 shows an example of a typical connection diagram for
driving the AD7961 using the two single-ended ADA4899-1
devices. The alternative ADC drivers are two single-ended
ADA4897-1 op amps or a differential amplifier ADA4932-1
that can drive the inputs of the AD7961.
If the analog inputs are multiplexed early enough, the inputs
can slew fast enough to a full-scale signal and settle the input
within the allowed time.
The AD7961 is an ideal fit for high speed multiplexed applications such as digital X-ray, computed tomography, and infrared
cameras that require superior performance in terms of noise,
power, and throughput, which significantly reduces cost in
these types of applications. The AD7961 has a quiet time
requirement of 90 ns to 110 ns during the conversion, where the
switching of multiplexer inputs (channels) must not occur to
avoid the corruption of conversion. In other words, a delay of
less than 90 ns and greater than 110 ns from the CNV± rising
The AD7961 offers extremely low noise floor relative to its fullscale input. The combination of high throughput rate, low noise
floor, and linearity also makes this part suitable for oversampling
applications such as spectroscopy, MRI gradient control, and
gas chromatography. The wide dynamic range of the AD7961
allows accurate measurements of both small and large signals
from multiple channels.
+VS
+5V
AD8031
ADR4550
+7V
0.1µF
10µF2
0.1µF
0.1µF
+5V
–VS
0.1µF
+1.8V
0.1µF
+1.8V
0.1µF
+VS
ADA4899-1
REF1
VDD1
VDD2
VIO
CNV±
100Ω
–VS
IN+
D±
100Ω
DCO±
100Ω
AD7961
IN–
+VS
33Ω
VCM = 2.5V
GND
VCM
56pF
100Ω
2.5V
0V TO 5 V
ADA4899-1
CLK±
DIGITAL HOST
LVDS TRANSMIT AND RECEIVE
REFIN
56pF
DIGITAL INTERFACE SIGNALS
33Ω
0V TO 5 V
VCM = 2.5V
0.1µF
–VS
+VS
VCM3
AD8031
0.1µF
–VS
1 SEE THE VOLTAGE REFERENCE OPTIONS SECTION. CONNECTION TO EXTERNAL REFERENCE SIGNALS IS DEPENDENT ON THE EN1
THE COMMON GROUND OF THE BOARD. THE REF AND REFIN PINS ARE DECOUPLED REGARDLESS OF EN1 AND EN0 SETTINGS.
3 BUFFERED VCM PIN OUTPUT GIVES THE REQUIRED 2.5V COMMON-MODE SUPPLY FOR ANALOG INPUTS.
Figure 32. Typical Application Diagram
Rev. B | Page 16 of 24
10888-015
AND EN0 SETTINGS.
2 A 10µF CAPACITOR WITH LOW ESL AND ESR IS USUALLY CONNECTED BETWEEN THE REF PIN AND REF_GND. CONNECT REF_GND TO
Data Sheet
AD7961
Table 8. Voltage Reference Options
EN3
X1
X1
EN2
0
0
EN1
0
0
EN0
0
1
REFIN
X1
0V
X1
0
0
1
2.048 V
X1
0
1
0
0V
X1
0
1
1
0V
0
1
X1
1
1
1
0
0
0
0
0
1
X1
X1
0V
X1
1
0
1
2.048 V
X1
1
1
0
0V
X1
1
1
1
0V
1
2
Reference Mode Description
Power-down mode. Everything is powered down, including the LVDS interface.
Interface powered up. Reference buffer disabled. An external 5 V reference is applied to the REF pin.
Connect REFIN to 0 V in this mode. The bandwidth of the input sampling network is set to 28 MHz.
Internal reference buffer enabled. An external 2.048 V reference applied to REFIN pin is required. A
buffered 4.096 V reference is available on the REF pin. The bandwidth of the input sampling
network is set to 28 MHz.
Internal reference buffer disabled. Drive the REF pins with a 4.096 V external reference. Connect
REFIN to 0 V in this mode. The bandwidth of the input sampling network is set to 28 MHz.
Snooze mode.2 LVDS powers down. The chip is unresponsive to CNV± start pulses. The wake-up
time is fast (5 μs) when EN3 to EN0 are set to XX01 or XX10. Ensure that the CNV± start pulse is low
when transitioning in and out of this mode.
Test patterns output on LVDS. The ADC output is not available on the interface.
Invalid mode.
Reference buffer disabled. Drive the REF pins with a 5 V external reference. The bandwidth of the
input sampling network is set to narrow (9 MHz).
Internal reference buffer enabled and driving REF pin to 4.096 V. The bandwidth of the input
sampling network is set to narrow (9 MHz).
Reference buffer disabled. Drive the REF pins with a 4.096 V external reference. The bandwidth of
the input sampling network is set to narrow (9 MHz).
Snooze mode.2 LVDS powers down. The chip is unresponsive to CNV± start pulses. The wake-up
time is fast (5 μs) when EN3 to EN0 are set to XX01 or XX10.
X = don’t care.
The snooze mode is not useful when the internal reference buffer is used because the fast wake-up is not possible due to the settling of the internal reference buffer.
VOLTAGE REFERENCE OPTIONS
Wake-Up Time from Power-Down and Snooze Modes
The AD7961 allows buffering of the reference voltage. The
AD7961 conversions are referred to a 5 V or 4.096 V reference
voltage. There are three options for using an external reference:
The AD7961 powers down when EN3 to EN0 = X000 and operates
in snooze mode when EN3 to EN0 = XX11 using the correct
reference choice as shown in Table 8. Typical wake-up times for
the selected reference settings from power-down and snooze
mode are shown in Table 9 and Table 10. Each wake-up time
represents the duration from the EN3 to EN0 logic transition to
when the ADC is ready for a CNV± rising edge. For example,
the user must wait 1.4 ms from power-down before applying
CNV± pulses to receive data conversion results when using
REFIN = 0 V.



Externally buffered reference source of 5 V applied to the
REF pin.
Externally buffered reference source of 4.096 V applied to
the REF pin.
External reference of 2.048 V applied to the REFIN pin
(high impedance input). The on-chip buffer gains this by 2
and drives the REF pin with 4.096 V.
The recommended external references for the AD7961 are the
ADR4520/ADR4540/ADR4550 and ADR440/ADR444/ADR445.
The various options for creating this reference are controlled by
the EN1 and EN0 pins (see Table 8). The −3 dB input bandwidth
is controlled by EN2. EN2 = 0 sets a −3 dB input bandwidth of
28 MHz, and EN2 = 1 sets a −3 dB input bandwidth of 9 MHz.
Use this lower bandwidth (9 MHz) only when the sample rate is
2 MSPS or lower. EN3 = 1 enables the VCM reference output,
and EN3 = 0 disables the VCM reference output voltage. The
best SNR and dynamic range performance is achieved by using
the larger 5 V external voltage reference option. The improvement
achieved is approximately 1.7 dB and is calculated using the
following equation:
Table 9. Wake-Up Time from Power-Down Mode, EN3 to
EN0 = X000
To Active Mode
EN3 to EN0 = XX01, REFIN = 0 V
EN3 to EN0 = XX01, REFIN = 2.048 V
EN3 to EN0 = XX10, REFIN = 0 V
Wake-Up Time
1.4 ms
8 ms
1.4 ms
Table 10. Wake-Up Time from Snooze Mode, EN3 to EN0 =
XX11
To Active Mode
EN3 to EN0 = XX01, REFIN = 0 V
EN3 to EN0 = XX01, REFIN = 2.048 V
EN3 to EN0 = XX10, REFIN = 0 V
5.0 
SNR  20 log

 4.096 
Rev. B | Page 17 of 24
Wake-Up Time
5 μs
8 ms
5 μs
AD7961
Data Sheet
POWER SUPPLY
Power-Up
The AD7961 uses both 5 V (VDD1) and 1.8 V (VDD2) power
supplies, as well as a digital input/output interface supply (VIO).
Drive the EN0 to EN3 pins with a 1.8 V logic level. VIO and
VDD2 can be taken from the same 1.8 V source; however, it is
best practice to isolate the VIO and VDD2 pins using separate
traces as well as to decouple each pin separately.
As is best practice for all ADCs, power on the core supplies
prior to applying an external reference (where applicable).
Apply the analog inputs last.
45
40
POWER DISSIPATION (mW)
The 5 V and 1.8 V supplies required for the AD7961 can be
generated using Analog Devices, Inc., LDOs such as the
ADP7104-5 and the ADP124-1.8. Figure 33 shows the PSRR vs.
supply frequency of the AD7961. The AD7961 core power
scales with throughput as shown in Figure 34, offering
significant power budget savings at lower speed operation.
When powering up the AD7961 device, first apply 1.8 V (VDD2,
VIO) to the device, then ramp 5 V (VDD1). Set the reference
configuration pins, EN0, EN1, and EN2, to the correct values.
When an internal reference buffer is used (governed by the EN1
and EN0 values), apply the external reference of 2.048 V to the
REFIN pin or 5 V/4.096 V to the REF pin.
110
VDD2 = 1.8V
VIO = 1.8V
VDD1 = 5.0V
100
80
30
25
20
15
10
10888-125
70
5
60
0
50
40
100
10888-121
PSRR (dB)
90
35
1k
10k
100k
1M
0
1
2
3
4
5
THROUGHPUT (MHz)
Figure 34. ADC Core Power Dissipation vs. Throughput, Self Clocked Mode,
CNV± in CMOS Mode, Internal Reference Buffer Disabled
FREQUENCY (Hz)
Figure 33. PSRR vs. Supply Frequency
Rev. B | Page 18 of 24
Data Sheet
AD7961
DIGITAL INTERFACE
The clock DCO± is a buffered copy of CLK± and is synchronous
to the data, D±, which is updated on the falling edge of DCO±
(tD). By maintaining good propagation delay matching between
D± and DCO± through the board and the digital host, DCO± can
be used to latch D± with good timing margin for the shift register.
CONVERSION CONTROL
All analog-to-digital conversions are controlled by the CNV±
signal. This signal can be applied in the form of a CNV+/CNV−
LVDS signal, or it can be applied in the form of a 1.8 V CMOS
logic signal to the CNV+ pin when CNV− is grounded. The
conversion is initiated by the rising edge of the CNV± signal.
Conversions are initiated by a rising edge of the CNV± pulse.
The CNV± pulse must be returned low (≤tCNVH maximum) for
valid operation. After a conversion begins, it continues until
completion. Additional CNV± pulses are ignored during the
conversion phase. After tMSB elapses, the host begins to burst the
CLK±. Note that tMSB is the maximum time for the MSB of the
new conversion result. Use tMSB as the gating device for CLK±.
The echoed clock, DCO±, and the data, D±, are driven in phase
with D± being updated on the falling edge of DCO±; the host
uses the rising edge of DCO± to capture D±. The only requirement is that the 16 CLK± pulses finish before tCLKL of the next
conversion phase elapses, or the data is lost. After all 16 bits are
read, up to tMSB, D± and DCO± are driven to 0. Set CLK± to idle
low between CLK± bursts.
After the AD7961 is powered up, the first conversion result
generated is valid. The key beneficial feature of the AD7961 is
that the user can return to the acquisition phase before the end
of the conversion.
The two methods for acquiring the digital data output of the
AD7961 via the LVDS interface are described in the Echoed
Clock Interface Mode and Self Clocked Mode sections.
Echoed Clock Interface Mode
The digital operation of the AD7961 in echoed clock interface
mode is shown in Figure 35. This interface mode, requiring
only a shift register on the digital host, can be used with many
digital hosts (such as FPGA, shift register, and microprocessor).
It requires three LVDS pairs (D±, CLK±, and DCO±) between
each AD7961 and the digital host.
SAMPLE N
SAMPLE N + 1
tCYC
tCNVH
CNV–
CNV+
tACQ
ACQUISITION
ACQUISITION
ACQUISITION
tCLKL
tCLK
15
CLK–
16
1
2
15
16
1
2
3
CLK+
tDCO
15
DCO–
16
1
2
15
1
16
2
3
DCO+
tMSB
D–
D1
N–1
D0
N–1
tD
0
D15
N
D14
N
D1
N
D0
N
Figure 35. Echoed Clock Interface Mode Timing Diagram
Rev. B | Page 19 of 24
0
D15
N+1
D14
N+1
D13
N+1
10888-018
tCLKD
D+
AD7961
Data Sheet
Self Clocked Mode
The self-clocked mode data capture method allows the digital
host to adapt its result capture timing to accommodate variations in propagation delay through any AD7961, for example,
where data is captured from multiple AD7961 devices sharing
a common input clock.
The digital operation of the AD7961 in self-clocked interface
mode is shown in Figure 36. This interface mode reduces the
number of traces between the ADC and the digital host to two
LVDS pairs (CLK± and D±) or to a single pair if sharing a
common CLK±. Multiple AD7961 devices can share a common
CLK± signal. This can be useful in reducing the number of
LVDS connections to the digital host.
Conversions are initiated by a CNV± pulse. The CNV± pulse
must be returned low (tCNVH maximum) for valid operation.
After a conversion begins, it continues until completion. Additional CNV± pulses are ignored during the conversion phase.
After the time, tMSB, elapses, the host begins to burst the CLK±
signal to the AD7961. All 18 CLK± pulses must be applied in
the window of time framed by tMSB and the subsequent tCLKL. The
required 18 CLK± pulses must finish before tCLKL (referenced to
the next conversion phase) elapses. Otherwise, the data is lost
because it is overwritten by the next conversion result.
When the self-clocked interface mode is used, each ADC
data-word is preceded by a 010 header sequence. After tMSB has
elapsed, the first bit of the header, 0, automatically appears on
D±, and the remaining two bits of the header, 10, are then
clocked out by the first two CLK± falling edges at the beginning
of the next sample. This header (010) is used to synchronize D±
of each conversion in the digital host because, in this mode,
there is no clock output synchronous to the data (D±) to allow
the digital host to acquire the data output.
Set CLK± to idle high between bursts of 18 CLK± pulses. The
header bit and conversion data of the next ADC result are
output on subsequent falling edges of CLK± during the next
burst of the CLK± signal.
Synchronization of the D± data to the acquisition clock of the
digital host is accomplished by using one state machine per
AD7961 device. For example, using a state machine that runs at
the same speed as CLK± incorporates three phases of this clock
frequency (120° apart). Each phase acquires the D± data as
output by the ADC.
When the self-clocked interface mode is used, the AD7961 also
allows the user to provide an extra (19th) clock pulse to see a
guaranteed 0 state at the end of the frame, as shown in Figure 37.
After tMSB has elapsed, the first bit of the header sequence, 0,
automatically appears on D± and the remaining two bits of the
header, 10, are then clocked out by the first two CLK± falling
edges at the beginning of the next sample. This header (010) is
used to synchronize D± of each conversion in the digital host
because, in this mode, there is no clock output synchronous to
the data (D±) to allow the digital host to acquire the data output.
The AD7961 data captured on each phase of the state
machine clock is then compared. The location of the 1 in
the header in each set of acquired data allows the user to
choose the state machine clock phase that occurs during
the data valid window of D±.
SAMPLE N
SAMPLE N + 1
tCYC
tCNVH
CNV–
CNV+
tACQ
ACQUISITION
ACQUISITION
ACQUISITION
tCLKL
tCLK
17
CLK–
18
1
2
4
3
17
1
18
2
3
CLK+
D+
D–
D1
N–1
D0
N–1
0
1
0
D15
N
D14
N
D1
N
Figure 36. Self Clocked Interface Mode Timing Diagram
Rev. B | Page 20 of 24
D0
N
0
1
0
D15
N+1
10888-019
tMSB
tCLKD
Data Sheet
AD7961
SAMPLE N
SAMPLE N + 1
tCYC
tCNVH
CNV–
CNV+
tACQ
ACQUISITION
ACQUISITION
ACQUISITION
tCLKL
tCLK
CLK–
19
20
21
1
2
4
3
17
18
1
19
2
3
CLK+
D+
D–
D1
N–1
D0
N–1
0
1
0
D15
N
D14
N
D1
N
D0
N
Figure 37. Self Clocked Interface Mode with Extra Clock Pulse Timing Diagram
Rev. B | Page 21 of 24
0
1
0
D15
N+1
10888-020
tMSB
tCLKD
AD7961
Data Sheet
APPLICATIONS INFORMATION
LAYOUT
Design the printed circuit board that houses the AD7961 so that
the analog and digital sections are separated and confined to
certain areas of the board. Avoid running digital lines under the
device because these couple noise onto the device, unless a
ground plane under the AD7961 is used as a shield. Do not run
fast switching signals, such as CNV± or CLK±, near analog
signal paths. Avoid crossover of digital and analog signals. Use
at least one ground plane. It can be common or split between
the digital and analog sections. In the latter case, join the planes
underneath the AD7961 devices.
The AD7961 voltage reference input pin, REF, has dynamic
input impedance. Decouple REF with minimal parasitic
inductances by placing the reference decoupling ceramic
capacitor close to and, ideally, right up against the REF and
REF_GND pins and connecting them with wide, low impedance
traces.
Finally, decouple the VDD1, VDD2, and VIO power supplies of
the AD7961 with ceramic capacitors, typically 100 nF, placed
close to the AD7961 and connected using short, wide traces to
provide low impedance paths and to reduce the effect of glitches
on the power supply lines.
EVALUATING AD7961 PERFORMANCE
Other recommended guidelines for the AD7961 schematic and
layout are outlined in the user guide of the EVAL-AD7961FMCZ
board (UG-581). The fully assembled and tested evaluation
board, user guide, and software for controlling the EVALAD7961FMCZ board from a PC via the EVAL-SDP-CH1Z are
available from the Analog Devices website at www.analog.com.
Rev. B | Page 22 of 24
Data Sheet
AD7961
OUTLINE DIMENSIONS
0.30
0.25
0.18
32
25
0.50
BSC
TOP VIEW
0.80
0.75
0.70
8
16
0.05 MAX
0.02 NOM
COPLANARITY
0.08
0.20 REF
SEATING
PLANE
3.25
3.10 SQ
2.95
EXPOSED
PAD
17
0.50
0.40
0.30
PIN 1
INDICATOR
1
24
9
BOTTOM VIEW
0.25 MIN
FOR PROPER CONNECTION OF
THE EXPOSED PAD, REFER TO
THE PIN CONFIGURATION AND
FUNCTION DESCRIPTIONS
SECTION OF THIS DATA SHEET.
COMPLIANT TO JEDEC STANDARDS MO-220-WHHD.
112408-A
PIN 1
INDICATOR
5.10
5.00 SQ
4.90
Figure 38. 32-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
5 mm × 5 mm Body, Very Very Thin Quad
(CP-32-7)
Dimensions shown in millimeters
ORDERING GUIDE
Model 1
AD7961BCPZ
AD7961BCPZ-RL7
EVAL-AD7961FMCZ
1
Temperature Range
−40°C to +85°C
−40°C to +85°C
Package Description
32-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
32-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
Evaluation Board
Z = RoHS Compliant Part.
Rev. B | Page 23 of 24
Package Option
CP-32-7
CP-32-7
AD7961
Data Sheet
NOTES
©2013–2014 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D10888-0-3/14(B)
Rev. B | Page 24 of 24