Throughput: 6 MSPS
SNR: 93 dB
INL: ±0.45 LSB typical, ±1 LSB maximum
DNL: ±0.3 LSB typical, ±0.5 LSB maximum
Power dissipation: 135 mW
32-lead LFCSP (5 mm × 5 mm)
SAR architecture
No latency/no pipeline delay
16-bit resolution with no missing codes
Zero error: ±1.5 LSB
Differential input voltage: ±4.096 V
Serial LVDS interface
Self-clocked mode
Echoed-clock mode
Can use LVDS or CMOS for conversion control (CNV signal)
Reference options
Internal: 4.096 V
External (1.2 V) buffered to 4.096 V
External: 4.096 V
APPLICATIONS
High dynamic range telecommunications
Receivers
Digital imaging systems
High speed data acquisition
Spectrum analysis
Test equipment
Differential ADC
AD7625
FUNCTIONAL BLOCK DIAGRAM
CAP
DAC
CM
÷2
SAR
CLOCK
Figure 1.
LOGIC
SERIAL
LVDS
VIO
CNV+, CNV–
D+, D–
DCO+, DCO–
CLK+, CLK–
REFIN REF
1.2V
BAND GAP
IN+
IN–
AD7625
GENERAL DESCRIPTION
The AD7625 is a 16-bit, 6 MSPS, charge redistribution successive
approximation register (SAR) based architecture analog-to-digital
converter (ADC). SAR architecture allows unmatched performance both in noise (93 dB SNR) and in linearity (1 LSB). The
AD7625 contains a high speed, 16-bit sampling ADC, an internal
conversion clock, and an internal buffered reference. On the
CNV± rising edge, it samples the voltage difference between the
IN+ and IN− pins. The voltages on these pins swing in opposite
phase between 0 V and REF. The 4.096 V reference voltage, REF,
can be generated internally or applied externally.
All converted results are available on a single LVDS self-clocked
or echoed-clock serial interface, reducing external hardware
connections.
The AD7625 is housed in a 32-lead, 5 mm × 5 mm LFCSP with
operation specified from −40°C to +85°C.
7652-001
Table 1. Fast PulSAR® ADC Selection
Input Type Resolution (Bits) 1 MSPS to <2 MSPS 2 MSPS to 3 MSPS 6 MSPS
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.
VDD1 = 5 V; VDD2 = 2.5 V; VIO = 2.5 V; REF = 4.096 V; all specifications T
Table 2.
Parameter Test Conditions/Comments Min Typ Max Unit
RESOLUTION 16 Bits
ANALOG INPUT
Voltage Range V
Operating Input Voltage V
− V
IN+
IN+
−V
IN−
, V
to GND −0.1 V
IN−
Common-Mode Input Range V
Common-Mode Rejection Ratio fIN = 1 MHz 60 dB
Input Current Midscale input 77 µA
THROUGHPUT
Complete Cycle 166 ns
Throughput Rate 0.1 6 MSPS
DC ACCURACY
Integral Linearity Error −1 ±0.45 +1 LSB
No Missing Codes 16 Bits
Differential Linearity Error −0.5 ±0.3 +0.5 LSB
Transition Noise 0.6 LSB
Zero Error T
MIN
to T
−4 ±1.5 +4 LSB
MAX
Zero Error Drift 0.5 ppm/°C
Gain Error T
MIN
to T
8 20 LSB
MAX
Gain Error Drift 0.4 ppm/°C
Power Supply Sensitivity
1
VDD1 = 5 V ± 5% 0.4 LSB
VDD2 = 2.5 V ± 5% 0.2 LSB
AC ACCURACY
External Reference fIN = 20 kHz
Dynamic Range 92.5 93.2 dB
Signal-to-Noise Ratio 92 93 dB
Spurious-Free Dynamic Range 106 dB
Total Harmonic Distortion −105.5 dB
Signal-to-(Noise + Distortion) 91.5 92 dB
Internal Reference fIN = 20 kHz
Dynamic Range 92.5 93.2 dB
Signal-to-Noise Ratio 91.5 92.9 dB
Spurious-Free Dynamic Range 106 dB
Total Harmonic Distortion −105.5 dB
Signal-to-(Noise + Distortion) 91 92.5 dB
−3 dB Input Bandwidth 100 MHz
Aperture Jitter 0.25 ps rms
INTERNAL REFERENCE
Output Voltage REFIN @ 25°C 1.2 V
Temperature Drift −40°C to +85°C ±15 ppm/°C
REFERENCE BUFFER
REFIN Input Voltage Range 1.2 V
REF Output Voltage Range 4.076 4.096 4.116 V
Line Regulation VDD1 ± 5%, VDD2 ± 5% 5 mV
EXTERNAL REFERENCE
Voltage Range REF 4.096 V
VCM PIN @ 25°C
Output Voltage REF/2 V
Output Impedance 4 5 6 kΩ
MIN
to T
REF
REF
, unless otherwise noted.
MAX
+V
/2 − 0.05 V
/2 V
REF
V
REF
+ 0.1 V
REF
/2 + 0.05 V
REF
Rev. 0 | Page 3 of 24
Page 4
AD7625
Parameter Test Conditions/Comments Min Typ Max Unit
LVDS I/O (ANSI-644)
Data Format Serial LVDS twos complement
Differential Output Voltage, VOD R
Common-Mode Output Voltage, V
OCM
Differential Input Voltage, VID 100 650 mV
Common-Mode Input Voltage, V
ICM
POWER SUPPLIES
Specified Performance
VDD1 4.75 5 5.25 V
VDD2 2.37 2.5 2.63 V
VIO 2.37 2.5 2.63 V
Energy per Conversion 6 MSPS throughput 22 nJ/sample
TEMPERATURE RANGE
Specified Performance T
1
Using an external reference.
2
The ANSI-644 LVDS specification has a minimum output common mode (V
3
Power dissipation is for the AD7625 device only. In self-clocked interface mode, 9 mW is dissipated in the 100 Ω terminator. In echoed-clock interface mode, 18 mW is
dissipated in two 100 Ω terminators.
= 100 Ω 200 350 454 mV
L
2
R
= 100 Ω 850 1250 1375 mV
L
800 1575 mV
Self-clocked mode and echoed-
11 13 mA
clock mode
Self-clocked mode and echoed-
13.5 16 mA
clock mode
Self-clocked mode and echoed-
13.5 16 mA
clock mode
to T
MIN
−40 +85 °C
MAX
) of 1125 mV.
OCM
Rev. 0 | Page 4 of 24
Page 5
AD7625
TIMING SPECIFICATIONS
VDD1 = 5 V; VDD2 = 2.5 V; VIO = 2.37 V to 2.63 V; REF = 4.096 V; all specifications T
Table 3.
Parameter Symbol Min Typ Max Unit
Time Between Conversions
Acquisition Time t
CNV± High Time t
CNV± to D± (MSB) Delay t
CNV± to Last CLK± (LSB) Delay t
CLK± Period
2
t
CLK± Frequency f
CLK± to DCO± Delay (Echoed-Clock Mode) t
DCO± to D± Delay (Echoed-Clock Mode) t
CLK± to D± Delay t
1
The maximum time between conversions is 10,000 ns. If CNV± is left idle for a time greater than the maximum value of t
2
For the minimum CLK period, the window available to read data is t
mode, n = 16; in self-clocked interface mode, n = 18.
1
t
CYC
ACQ
CNVH
MSB
CLKL
CLK
CLK
DCO
D
CLKD
CYC
− t
166 10,000 ns
40 ns
10 40 ns
145 ns
110 ns
(t
− t
CYC
MSB
250 300 MHz
0 4 7 ns
0 1 ns
0 4 7 ns
+ t
. Divide this time by the number of bits (n) that are read. In echoed-clock interface
MSB
CLK
to T
MIN
+ t
)/n 4 3.33 ns
CLK
, unless otherwise noted.
MAX
, the subsequent conversion result is invalid.
CYC
Rev. 0 | Page 5 of 24
Page 6
AD7625
ABSOLUTE MAXIMUM RATINGS
Table 4.
Parameter Rating
Analog Inputs/Outputs
IN+, IN− to GND1
−0.3 V to REF + 0.3 V or
±130 mA
REF2 to GND −0.3 V to +6 V
VCM, CAP2 to GND −0.3 V to +6 V
CAP1, REFIN to GND −0.3 V to +2.7 V
Supply Voltage
VDD1 −0.3 V to +6 V
VDD2, VIO −0.3 V to +3 V
Digital Inputs to GND −0.3 V to VIO + 0.3 V
Digital Outputs to GND −0.3 V to VIO + 0.3 V
Input Current to Any Pin Except
Supplies
3
Operating Temperature Range
±10 mA
−40°C to +85°C
(Commercial)
Storage Temperature Range −65°C to +150°C
Junction Temperature 150°C
ESD 1 kV
1
See the Analog Inputs section.
2
Keep CNV+/CNV− low for any external REF voltage > 4.3 V applied
to the REF pin.
3
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.
THERMAL RESISTANCE
θJA is specified for the worst-case conditions, that is, a device
soldered in a circuit board for surface-mount packages.
Table 5. Thermal Resistance
Package Type θJA θ
Unit
JC
32-Lead LFCSP_VQ 40 4 °C/W
ESD CAUTION
Rev. 0 | Page 6 of 24
Page 7
AD7625
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
P2
F
REF
GND
RE
REF
CA
GND
CAP2
CAP2
29
28
27
26
31
30
32
25
VDD1
1
2
3
4
5
6
7
8
PIN 1
INDICATOR
AD7625
TOP VIEW
(Not to Scale)
9
11
10
12
D–
D+
VIO
CNV+
13
14
15
GND
DCO–
DCO+
VDD2
CAP1
REFIN
EN0
EN1
VDD2
CNV–
NOTES
1. CONNECT T HE EXPOSED P AD TO THE G ROUND
PLANE OF T HE PCB USING MULTIPLE VIAS.
GND
24
IN+
23
IN–
22
VCM
21
VDD1
20
VDD1
19
VDD2
18
CLK+
17
16
CLK–
07652-002
Figure 2.
Table 6. Pin Function Descriptions
Pin No. Mnemonic Type
1
Description
1 VDD1 P Analog 5 V Supply. Decouple the 5 V supply with a 100 nF capacitor.
2 VDD2 P
Analog 2.5 V Supply. Decouple this pin with a 100 nF capacitor. The 2.5 V supply source should
supply this pin first and then be traced to the other VDD2 pins (Pin 7 and Pin 18).
3 CAP1 AO Connect this pin to a 10 nF capacitor.
4 REFIN AI/O
Prebuffer Reference Voltage. When using the internal reference, this pin outputs the band gap voltage
and is nominally at 1.2 V. It can be overdriven with an external reference voltage such as the ADR280.
In either internal or external reference mode, a 10 F capacitor is required. If using an external 4.096 V
reference (connected to REF), this pin is a no connect and does not require any capacitor.
5, 6 EN0, EN1 DI Enable Pins. The logic levels of these pins set the operation of the device as follows:
EN1 = 0, EN0 = 0: Illegal state.
EN1 = 0, EN0 = 1: Enable internal buffer, disable internal reference. An external 1.2 V reference
connected to the REFIN pin is required.
EN1 = 1, EN0 = 0: Disable internal reference and reference buffer. An external 4.096 V reference
connected to the REF pin is required.
EN1 = 1, EN0 = 1: Enable internal reference and reference buffer.
7 VDD2 P Digital 2.5 V Supply. Decouple this pin with a 100 nF capacitor.
8, 9 CNV−, CNV+ DI
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.
10, 11 D−, D+ DO LVDS Data Outputs. The conversion data is output serially on these pins.
12 VIO P Input/Output Interface Supply. Use a 2.5 V supply and decouple this pin with a 100 nF capacitor.
13 GND P Ground. Return path for the 100 nF capacitor connected to Pin 12.
14, 15 DCO−, DCO+ DO
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 a 2-bit header (10) to allow synchronization of
the data by the digital host with simple logic. When DCO+ is not grounded, the echoed-clock inter-
face 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 latched in the digital host on the next rising edge of DCO+.
16, 17 CLK−, CLK+ DI LVDS Clock Inputs. This clock shifts out the conversion results on the falling edge of CLK+.
18 VDD2 P Analog 2.5 V Supply. Decouple this pin with a 100 nF capacitor.
19, 20 VDD1 P
Analog 5 V Supply. Isolate these pins from Pin 1 with a ferrite bead and decouple them with a 100 nF
capacitor.
21 VCM AO
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.
22 IN− AI Differential Negative Analog Input. Referenced to and must be driven 180° out of phase with IN+.
23 IN+ AI Differential Positive Analog Input. Referenced to and must be driven 180° out of phase with IN−.
24 GND P Ground.
Rev. 0 | Page 7 of 24
Page 8
AD7625
Pin No. Mnemonic Type
25, 26, 28 CAP2 AO
1
Description
Connect all three CAP2 pins together and decouple them with the shortest trace possible to a single
10 F, low ESR, low ESL capacitor. The other side of the capacitor must be placed close to Pin 27 (GND).
27 GND P Ground. Return path for the 10 F capacitor connected to Pin 25, Pin 26, and Pin 28.
29, 30, 32 REF AI/O
Buffered Reference Voltage. When using the internal reference or the 1.2 V external reference (REFIN
input), the 4.096 V system reference is produced at this pin. When using an external reference, such
as the ADR434 or the ADR444, the internal reference buffer must be disabled. In either case, connect
all three REF pins together and decouple them with the shortest trace possible to a single 10 F, low
ESR, low ESL capacitor. The other side of the capacitor must be placed close to Pin 31 (GND).
31 GND P Ground. Return path for the 10 F capacitor connected to Pin 29, Pin 30, and Pin 32.
EP 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. See the Exposed Paddle section for more information.
1
AI = analog input; AI/O = bidirectional analog; AO = analog output; DI = digital input; DO = digital output; P = power.
Rev. 0 | Page 8 of 24
Page 9
AD7625
TYPICAL PERFORMANCE CHARACTERISTICS
0
–20
–40
–60
–80
–100
AMPLI TUDE (d B)
–120
–140
–160
–180
00.51.01.52.02.53.0
FREQUENCY (MHz )
Figure 3. FFT 2 kHz Input Tone, Full View
07652-005
0
–20
–40
–60
–80
–100
AMPLITUDE (dB)
–120
–140
–160
–180
051015202530354045
FREQUENCY (kHz)
INPUT TO NE = 2kHz
SNR = 93.16dB
SINAD = 92.09d B
THD = –110.45dB
SFDR = 111.37d B
Figure 6. FFT 2 kHz Input Tone, Zoom In on Input Tone and Harmonics
07652-006
0
–20
–40
–60
–80
–100
AMPLITUDE (dB)
–120
–140
–160
–180
00. 51.01.52.02.53. 0
FREQUENCY (MHz)
INPUT TO NE = 50kHz
SNR = 93.04dB
SINAD = 92.63dB
THD = –103.57d B
SFDR = –102.69dB
Figure 4. FFT 50 kHz Input Tone
0.5
0.4
0.3
0.2
0.1
0
DNL (LSB)
–0.1
–0.2
–0.3
–0.4
–0.5
016,38432,76849,15265,536
CODE
Figure 5. Differential Nonlinearity vs. Code
07652-007
07652-015
0
–20
–40
–60
–80
–100
AMPLITUDE (dB)
–120
–140
–160
–180
00. 51.01.52.02.53.0
FREQUENCY (MHz)
INPUT TO NE = 100kHz
SNR = 92.91dB
SINAD = 92.55dB
THD = –103.11d B
SFDR = –103.41dB
Figure 7. FFT 100 kHz Input Tone
1.0
0.8
0.6
0.4
0.2
0
INL (LSB)
–0.2
–0.4
–0.6
–0.8
–1.0
016,38432,76849, 15265,536
CODE
Figure 8. Integral Nonlinearity vs. Code
07652-008
07652-014
Rev. 0 | Page 9 of 24
Page 10
AD7625
–
–
94
–96
–98
–100
–102
–104
–106
THD (dB)
–108
–110
–112
–114
–116
020406080100120
–1dBFS
INPUT FREQ UENCY (kHz)
–0.5dBFS
–3dBFS
–10dBFS
–5dBFS
07652-012
Figure 9. THD at Input Amplitudes of −0.5 dBFS to −10 dBFS vs. Frequency
100.5
–101.0
–101.5
–102.0
THD (dB)
–102.5
–103.0
–103.5
–60 –40–20020406080100 120
INTERNAL REF
EXTERNAL REF
TEMPERATURE ( °C)
Figure 12. THD vs. Temperature (−0.5 dB, 20 kHz Input Tone)
07652-021
93.8
93.6
93.4
93.2
93.0
92.8
92.6
SNR, DYNAMIC RANGE ( dB)
92.4
92.2
SNR vs. TEMP EXTERNAL REF
SNR vs. TEMP I NTERNAL REF
–60 –40–20020406080100 120
DYNR vs. TEMP I NTERNAL REF
DYNR vs. TEMP EXTERNAL REF
TEMPERATURE ( °C)
Figure 10. Dynamic Range and SNR vs. Temperature
(−0.5 dB, 20 kHz Input Tone)
120
100
80
60
40
20
0
INPUT CURRENT (µ A)
–20
–40
–60
–4–6–20246
DIFFERENTIAL INPUT VOLTAGE (V)
IN+
IN–
Figure 11. Input Current (IN+, IN−) vs. Differential Input Voltage (6 MSPS)
93.2
93.0
92.8
92.6
92.4
SINAD (dB)
92.2
92.0
91.8
91.6
07652-018
SINAD vs. TEMP INTERNAL RE F
–60 –40–20020406080100 120
SINAD vs. TEMP EXTERNAL REF
TEMPERATURE (° C)
07652-019
Figure 13. SINAD vs. Temperature
(−0.5 dB, 20 kHz Input Tone)
12
10
8
6
4
2
ZERO ERROR AND GAIN ERROR (LSB)
0
–60 –40–20020406080100 120
07652-010
GAIN ERROR
ZERO ERROR
TEMPERATURE (° C)
07652-020
Figure 14. Zero Error and Gain Error vs. Temperature
Rev. 0 | Page 10 of 24
Page 11
AD7625
250,000
200,000
150,000
262,144 SAMPLES
STD DEVIATI ON = 0.4829
201,320
140,000
120,000
100,000
80,000
128,084
129,601
262,144 SAMPLES
STD DEVIATI ON = 0 .5329
COUNT
100,000
50,000
0
0
FEC7 FEC8FEC9 FECA F ECB FECC F ECD
30,651
54
CODE (HEX)
Figure 15. Histogram of 262,144 Conversions of a DC Input
at the Code Center (Internal Reference)
250,000
200,000
150,000
COUNT
100,000
50,000
41
0
0
FEC8FEC9 FECA FECB FECC FECD FECE
201,614
30,206
CODE (HE X)
Figure 16. Histogram of 262,144 Conversions of a DC Input
at the Code Center (External Reference)
30,073
46
262,144 SAMPLES
STD DEVIATION = 0.4814
30,250
33
60,000
COUNT
40,000
20,000
2130
0
07652-022
00
0
FEC6FEC7FEC8FEC9FECAFECB
CODE (HEX)
2329
07652-023
Figure 17. Histogram of 262,144 Conversions of a DC Input
at the Code Transition (Internal Reference)
0
07652-024
Rev. 0 | Page 11 of 24
Page 12
AD7625
TERMINOLOGY
Common-Mode Rejection Ratio (CMRR)
CMRR is defined as the ratio of the power in the ADC output at
full-scale frequency, f, to the power of an 80 mV p-p sine wave
applied to the common-mode voltage of V
frequency f
.
S
CMRR (dB) = 10log(Pf/Pf
)
S
IN+
and V
IN−
at
where:
Pf is the power at frequency f in the ADC output.
Pf
is the power at frequency fS in the ADC output.
S
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 = [(SINAD
− 1.76)/6.02]
dB
Gain Error
The first transition (from 100 … 000 to 100 …001) should occur
at a level ½ LSB above nominal negative full scale (−4.0959375 V
for the ±4.096 V range). The last transition (from 011 … 110 to
011 … 111) should occur for an analog voltage 1½ LSB below
the nominal full scale (+4.0959375 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.
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
V
LSB2(V) =
INp-p
N
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.
Reference Voltage Temperature Coefficient
The reference voltage temperature coefficient is derived from the
typical shift of output voltage at 25°C on a sample of parts at the
maximum and minimum reference output voltage (V
ured at T
, T(25°C), and T
MIN
REF
Cppm/×
=°
)(TCV
. It is expressed in ppm/°C as
MAX
((
REFREF
C25
REF
×°
MAX
MIN
) meas-
REF
)MinV–)MaxV
6
10
)T–T()(V
where:
(Max) = maximum V
V
REF
(Min) = minimum V
V
REF
V
(25°C) = V
REF
T
MAX
T
MIN
= +85°C.
= −40°C.
REF
at 25°C.
REF
REF
at T
at T
MIN
MIN
, T(25°C), or T
, T(25°C), or T
MAX
MAX
.
.
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 + 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.
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.
Rev. 0 | Page 12 of 24
Page 13
AD7625
THEORY OF OPERATION
IN+
MSB
REF
(4.096V)
GND
IN–
32,768C 16,384C4C2CCC
32,768C 16,384C4C2CCC
MSB
Figure 18. ADC Simplified Schematic
CIRCUIT INFORMATION
The AD7625 is a 6 MSPS, high precision, power efficient, 16-bit
ADC that uses SAR based architecture to provide performance
of 93 dB SNR, ±0.45 LSB INL, and ±0.3 LSB DNL.
The AD7625 is capable of converting 6,000,000 samples per
second (6 MSPS). The device typically consumes 135 mW. The
AD7625 offers the added functionality of a high performance
on-chip reference and on-chip reference buffer.
The AD7625 is specified for use with 5 V and 2.5 V supplies
(VDD1, VDD2). The interface from the digital host to the
AD7625 uses 2.5 V logic only. The AD7625 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 sou r c e .
The AD7625 is housed in a space-saving, 32-lead, 5 mm ×
5 mm LFCSP.
CONVERTER INFORMATION
The AD7625 is a 6 MSPS ADC that uses SAR based architecture
incorporating a charge redistribution DAC. Figure 18 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 logic high. Note
that the AD7625 can receive a CMOS (CNV+) or LVDS format
(CNV±) signal.
GND
LSB
LSB
SWITCHES
COMP
CONTRO L
CONTROL
CNV+, CNV–
CONVERSION
CONTROL
LOGIC
CLK+, CLK–
DCO+, DCO–
D+, D–
OUTPUT CODE
LVDS INTERF ACE
DATA TRANSFER
SW+
SW–
GND
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 4.096 V
(the reference voltage), the comparator input varies by binary
weighted voltage steps (V
REF
/4 … V
REF
/65,536). The
REF
/2, V
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 AD7625 digital interface uses low voltage differential
signaling (LVDS) to enable high data transfer rates.
The AD7625 conversion result is available for reading after t
(time from the conversion start until MSB is available) has
elapsed. The user must apply a burst LVDS CLK± signal to the
AD7625 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 is illustrated
in Figure 29 and Figure 30 and is characterized as follows: The
differential voltage on CLK± should be held to create logic low
in the time between t
CLKL
and t
MSB
.
The AD7625 has two data read modes. For more information
about the echoed-clock and self-clocked interface modes, see
the Digital Interface section.
07652-030
MSB
Rev. 0 | Page 13 of 24
Page 14
AD7625
V
TRANSFER FUNCTIONS
The AD7625 uses a 4.096 V reference. The AD7625 converts
the differential voltage of the antiphase analog inputs (IN+ and
IN−) into a digital output. The analog inputs, IN+ and IN−,
require a 2.048 V common-mode voltage (REF/2).
The 16-bit conversion result is in MSB first, twos complement
format.
The ideal transfer functions for the AD7625 are shown in
Figure 19 and Table 7.
011 ... 111
011 ... 110
011 ... 101
ADC CODE (TWO S COMPLEMENT)
100 ... 010
100 ... 001
100 ... 000
–FSR
–FSR + 1LSB
–FSR + 0.5L SB
+FSR – 1.5L SB
ANALOG INPUT
Figure 19. ADC Ideal Transfer Functions (FSR = Full-Scale Range)
The analog inputs, IN+ and IN−, applied to the AD7625 must be
180° out of phase with each other. Figure 20 shows an equivalent
circuit of the input structure of the AD7625.
The two diodes provide ESD protection for the analog inputs,
IN+ and IN−. Care must be taken to ensure that the analog input
signal does not exceed the reference voltage by more than 0.3 V.
If the analog input signal exceeds this level, the diodes become
forward-biased and start conducting current. These diodes can
handle a forward-biased current of 130 mA maximum. However,
if the supplies of the input buffer (for example, the supplies of
the ADA4899-1 in Figure 24) are different from those of the
reference, 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.
DD1
IN+
OR
IN–
Figure 20. Equivalent Analog Input Circuit
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
AD7625 shows some degradation in THD with higher analog
input frequencies.
85
80
75
70
65
60
CMRR (dB)
55
50
45
40
1101001k10k100k1M10M
INPUT COMMON-MODE FREQUENCY (Hz)
Figure 21. Analog Input CMRR vs. Frequency
250Ω
CNV
25pF
7652-032
07652-009
Rev. 0 | Page 14 of 24
Page 15
AD7625
TYPICAL CONNECTION DIAGRAM
VDD2
(2.5V)
100nF
8
ADR280
VIO
CONTROL F OR
ENABLE
VDD2
(2.5V)
V+
VDD1
(5V)
100nF
10nF
10µF
10kΩ
PINS
100nF
CONVERS ION
CONTROL
CMOS (CNV+ ONLY)
LVDS CNV+ AND CNV–
USING 100Ω
TERMINATION RESI STOR
8
ADR434
ADR444
CAPACITOR ON OUTPUT
FOR STABILITY
1
VDD1
2
VDD2
3
CAP1
4
3
10kΩ
OR
REFIN
5
EN0
6
EN1
7
VDD2
4
891011121314151617
C
REF
1, 2
10µF
3231302928272625
REF
REF
GND
REF
PADDLE
AD7625
CNV–
CNV+
D–
D+
VIO
100Ω
100Ω
VIO
(2.5V)
1
10µF
GND
CAP2
GND
CAP2
CAP2
DCO–
DCO+
5
CLK–
GND
IN+
IN–
VCM
VDD1
VDD1
VDD2
24
23
22
21
20
19
18
100nF
IN+
SEE THE DRIVING
IN–
THE AD7625 SECTI ON
VCM
FERRITE
BEAD
100nF
6
VDD2
(2.5V)
VDD1
(5V)
7
CLK+
100Ω
100Ω
DIGITAL INTERF ACE SIGNALS
DIGITAL HOS T
LVDS TRANSMIT AND RECEIVE
1
SEE THE LAYOUT, DECOUPLING, AND G ROUNDING SECT ION.
2
C
IS USU ALLY A 10µF CERAMIC CAPACIT OR
REF
3
USE PULL-UP OR PULL-DOWN RESIS TORS TO CONTROL EN0, EN1 DURING P OWER-UP. EN0 AND EN1 INPUTS CAN BE
FIXED IN HARDWARE OR CONTRO LLED USI NG A DIGIT AL HOST (EN0 = 0 AND EN1 = 0 IS AN I LLEGAL STATE).
4
OPTIO N TO USE A CMOS (CNV+) OR LV DS (CNV±) INPUT TO CONTRO L CONVERSI ONS.
5
TO ENABLE SELF-CLO CKED MODE, T IE DCO+ TO GND USING A PULL-DOWN RESISTOR.
6
CONNECT PIN 19 AND PIN 20 TO VDD1 SUPPLY; ISOLATE FROM PIN 1 USING A FERRI TE BEAD SIM ILAR TO WURTH 74279266.
7
SEE THE DRIVING THE AD7625 S ECTION FOR DETAIL S ON AMPLI FIER CONFIGURATI ONS.
8
SEE THE VO LTAGE REF ERENCE OPTI ONS SECTI ON FOR DET AILS.
WITH LOW ESR AND ESL.
Figure 22. Typical Application Diagram
07652-027
Rev. 0 | Page 15 of 24
Page 16
AD7625
DRIVING THE AD7625
Differential Analog Input Source
Figure 24 shows an ADA4899-1 driving each differential input
to the AD7625.
Single-Ended-to-Differential Driver
For applications using unipolar analog signals, a single-endedto-differential driver, as shown in Figure 23, allows for a differential input into the part. This configuration, when provided
with an input signal of 0 V to 4.096 V, produces a differential
±4.096 V with midscale at 2.048 V. The one-pole filter using
R = 33 Ω and C = 56 pF provides a corner frequency of 86 MHz.
The VCM output of the AD7625 can be buffered and then used
to provide the required 2.048 V common-mode voltage.
BUFFERED VCM PI N OUTPUT
GIVES T HE REQUIRE D 2.048V
COMMON-MODE SUPPLY FOR
ANALOG INPUT S.
1
SEE THE VOL TAGE REFERENCE OPTI ONS SECTI ON. CONNECTION TO EXTERNAL REFERENCE SIGNALS
IS DEPENDENT ON THE EN1 AND EN0 SE TTINGS.
2
C
IS USUALL Y A 10µF CERAMIC CAPACI TOR
REF
THE REF AND REF IN PINS ARE DECO UPLED REGARDL ESS OF E N1 AND EN0 SETTI NGS.
–V
+V
–V
56pF
S
S
33Ω
56pF
S
+V
S
0.1µF
AD8031, AD8032
–V
S
WITH LOW ESL AND ESR.
IN+
IN–
C
REF
10µF
2
REF
AD7625
GND
REFIN
VCM
2.048V
C
REF
10µF
2
REF
1
07652-025
Figure 24. Driving the AD7625 from a Differential Analog Source
Rev. 0 | Page 16 of 24
Page 17
AD7625
VOLTAGE REFERENCE OPTIONS
The AD7625 allows flexible options for creating and buffering
the reference voltage. The AD7625 conversions refer to 4.096 V
only. The various options creating this 4.096 V reference are
controlled by the EN1 and EN0 pins (see Tabl e 8).
A
V+
Table 8. Voltage Reference Options1
Option EN1 EN0 Reference Mode
A 1 1
Use internal reference and internal
reference buffer (both are enabled).
B 0 1
Use external 1.2 V reference with
internal reference buffer enabled.
The internal reference is disabled.
C 1 0
Use external 4.096 V reference with
an external reference buffer. The
internal reference and reference
buffer are disabled.
1
EN1 = 0 and EN0 = 0 is an illegal state.
SETTING EN1 = 1 AND EN0 = 1 ENABLES THE INT ERNAL
REFERENCE AND REFERENCE BUFFER. DECOUPLE
THE REF AND REFI N PINS EXTERNALLY.
V+
DISABLES THE INTERNAL REF ERENCE
AND REFERENCE BUFFER. CONNECT THE
ADR434
ADR444
V–
SETTING EN1 = 1 AND EN0 = 0
BUFFERED 4.096V SIGNAL T O
THE REF PIN.
10µF
C
REF
IN+
10µF
REFIN
AD7625
IN–
ADR280
B
SETTING EN1 = 0 AND EN0 = 1
DISABLES THE INTERNAL REF ERENCE
AND ENABLES THE INTERNAL REFERE NCE BUFFER.
CONNECT A 1.2V REFERENCE TO THE REFIN P IN. THE 1. 2V
APPLIED TO THE REFI N PIN IS BUFFERED INTERNAL LY
TO CREATE A 4.096V REFERENCE FOR THE ADC.
V+
7652-026
Figure 25. Voltage Reference Options
Rev. 0 | Page 17 of 24
Page 18
AD7625
POWER SUPPLY
The AD7625 uses both 5 V (VDD1) and 2.5 V (VDD2) power
supplies, as well as a digital input/output interface supply (VIO).
VIO allows a direct interface with 2.5 V logic only. VIO and
VDD2 can be taken from the same 2.5 V source; however, it is
best practice to isolate the VIO and VDD2 pins using separate
traces and also to decouple each pin separately.
The 5 V and 2.5 V supplies required for the AD7625 can be
generated using Analog Devices, Inc., low dropout regulators
(LDOs) such as the ADP3330-2.5, ADP3330-5, ADP3334, and
ADP1708.
90
VDD2
85
80
VDD1
75
70
PSRR (dB)
65
60
55
50
1101001k10k
SUPPLY FREQ UENCY (Hz)
Figure 26. PSRR vs. Supply Frequency
(350 mV pp Ripple on VDD2, 600 mV Ripple on VDD1)
Power-Up
When powering up the AD7625 device, first apply the VIO
voltage to the device so that the EN1 and EN0 values can be set
for the reference option in use. Connect the EN0 and EN1 pins
to pull-up/pull-down resistors to ensure that one or both of
these pins is set to a nonzero value. EN0 = 0 and EN1 = 0 is an
illegal state that must be avoided.
INTERNAL REFERENCE USED
07652-011
After VIO is established, apply the 2.5 V VDD2 supply to the
device followed by the 5 V VDD1 supply and then an external
reference (depending on the reference setting being used).
Finally, apply the analog inputs to the ADC.
25
20
15
10
CURRENT (mA)
5
0
01000200030004000500060007000
VDD2 INTERNAL REF
VDD2 EXTERNAL REF
VIO
VDD1 INTERNAL REF
VDD1 EXTERNAL REF
SAMPLING RAT E (kSPS)
Figure 27. Current Consumption vs. Sampling Rate
150
140
130
120
110
100
90
POWER DISSIPATIO N (mW)
80
70
60
01000200030004000500060007000
INTERNAL REF
EXTERNAL REF
SAMPLING RAT E (kSPS)
Figure 28. Power Dissipation vs. Sampling Rate
07652-016
07652-017
Rev. 0 | Page 18 of 24
Page 19
AD7625
A
DIGITAL INTERFACE
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 2.5 V CMOS
logic signal to the CNV+ pin. The conversion is initiated by the
rising edge of the CNV signal.
After the AD7625 is powered up, the first conversion result
generated is invalid. Subsequent conversion results are valid
provided that the time between conversions does not exceed
the maximum specification for t
The two methods for acquiring the digital data output of the
AD7625 via the LVDS interface are described in the following
sections.
Echoed-Clock Interface Mode
The digital operation of the AD7625 in echoed-clock interface
mode is shown in Figure 29. This interface mode, requiring
only a shift register on the digital host, can be used with many
digital hosts (FPGA, shift register, microprocessor, and so on).
It requires three LVDS pairs (D±, CLK±, and DCO±) between
each AD7625 and the digital host.
CNV–
CNV+
.
CYC
SAMPLE NS
t
CNVH
t
CYC
t
ACQ
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+
(t
). By maintaining good propagation delay matching between
D
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.
Conversions are initiated by a CNV± pulse. The CNV± pulse
must be returned low (≤t
maximum) for valid operation.
CNVH
After a conversion begins, it continues until completion. Additional CNV± pulses are ignored during the conversion phase.
After the time t
CLK±. Note that t
elapses, the host should begin to burst the
MSB
is the maximum time for the MSB of the
MSB
new conversion result and should be used 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 should use the rising edge of DCO+ to capture
D±. The only requirement is that the 16 CLK pulses finish
before the time t
data is lost. From the time t
elapses for the next conversion phase or the
CLKL
to t
CLKL
, D± and DCO± are
MSB
driven to 0s. Set CLK± to idle low between CLK± bursts.
The digital operation of the AD7625 in self-clocked interface
mode is shown in Figure 30. This interface mode reduces the
number of wires between ADCs and the digital host to two LVDS
pairs per AD7625 (CLK± and D±) or to a single pair if sharing a
common CLK± using multiple AD7625 devices. Self-clocked
interface mode facilitates the design of boards that use multiple
AD7625 devices. The digital host can adapt the interfacing
scheme to account for differing propagation delays between
each AD7625 device and the digital host.
The self-clocked interface mode consists of preceding the results
of each ADC word with a 2-bit header on the data, D±. This
header is used to synchronize D± of each conversion in the
digital host. Synchronization is accomplished by one simple
state machine per AD7625 device. For example, if the state
machine is running at the same speed as CLK± with three
phases, the state machine measures when the Logic 1 of the
header occurs.
CNV–
CNV+
SAMPLE N
t
CNVH
t
CYC
t
ACQ
Conversions are initiated by a CNV± pulse. The CNV± pulse
must be returned low (≤t
maximum) for valid operation.
CNVH
After a conversion begins, it continues until completion. Additional CNV± pulses are ignored during the conversion phase.
After the time t
CLK±. Note that t
elapses, the host should begin to burst the
MSB
is the maximum time for the first bit of
MSB
the header and should be used as the gating device for CLK±.
CLK± is also used internally on the host to begin the internal
synchronization state machine. The next header bit and conversion
results are output on subsequent falling edges of CLK±. The
only requirement is that the 18 CLK± pulses finish before the
time t
elapses for the next conversion phase or the data is
CLKL
lost. Set CLK± to idle high between bursts of 18 CLK± pulses.
When laying out the printed circuit board (PCB) for the AD7625,
follow the practices described in this section to obtain the maximum performance from the converter.
Exposed Paddle
The AD7625 has an exposed paddle on the underside of the
package.
• Solder the paddle directly to the PCB.
• Connect the paddle to the ground plane of the board using
multiple vias, as shown in Figure 31.
• Decouple all supply pins except for Pin 12 (VIO) directly to
the paddle, minimizing the current return path.
• Pin 13 and Pin 24 can be connected directly to the paddle.
Use vias to ground at the point where these pins connect to
the paddle.
VDD1 Supply Routing and Decoupling
The VDD1 supply is connected to Pin 1, Pin 19, and Pin 20. The
supply should be decoupled using a 100 nF capacitor at Pin 1.
The user can connect this supply trace to Pin 19 and Pin 20. Use
a series ferrite bead to connect the VDD1 supply from Pin 1 to
Pin 19 and Pin 20. The ferrite bead isolates any high frequency
noise or ringing on the VDD1 supply. Decouple the VDD1
supply to Pin 19 and Pin 20 using a 100 nF capacitor to GND.
This GND connection can be placed a short distance away from
the exposed paddle.
VIO Supply Decoupling
Decouple the VIO supply applied to Pin 12 to ground at Pin 13.
Layout and Decoupling of Pin 25 to Pin 32
Connect the outputs of Pin 25, Pin 26, and Pin 28 together and
decouple them to Pin 27 using a 10 μF capacitor with low ESR
and low ESL.
Reduce the inductance of the path connecting Pin 25, Pin 26,
and Pin 28 by widening the PCB traces connecting these pins.
A similar approach should be taken in the connections used for
the reference pins of the AD7625. Connect Pin 29, Pin 30, and
Pin 32 together using widened PCB traces to reduce inductance.
In internal or external reference mode, a 4.096 V reference voltage
is output on Pin 29, Pin 30, and Pin 32. Decouple these pins to
Pin 31 using a 10 μF capacitor with low ESR and low ESL.
Figure 31 shows an example of the recommended layout for
the underside of the AD7625 device. Note the extended signal
trace connections and the outline of the capacitors decoupling
the signals applied to the REF pins (Pin 29, Pin 30, and Pin 32)
and to the CAP2 pins (Pin 25, Pin 26, and Pin 28).
24 23 22 21 20 19 18 17
4.096V
EXTERNAL REFERENCE
(ADR434 OR ADR444)
Figure 31. PCB Layout and Decoupling Recommendations for Pin 24 to Pin 32
25
26
27
28
29
30
31
32
Paddle
12345678
16
15
14
13
12
11
10
9
07652-013
Rev. 0 | Page 21 of 24
Page 22
AD7625
C
OUTLINE DIMENSIONS
0.08
0.60 MAX
25
24
EXPOSED
PAD
(BOTTOM VIEW)
17
16
3.50 REF
FOR PROPER CONNECTION O F
THE EXPOSED PAD, REFER TO
THE PIN CONF IGURATIO N AND
FUNCTION DESCRIPTIONS
SECTION OF THIS DATA SHEET.
PIN 1
32
9
INDICATOR
1
3.25
3.10 SQ
2.95
8
0.25 MIN
011708-A
5.00
INDI
1.00
0.85
0.80
PIN 1
ATO R
12° MAX
SEATING
PLANE
BSC SQ
TOP
VIEW
0.80 MAX
0.65 TYP
0.30
0.23
0.18
COMPLIANT TO JEDEC STANDARDS MO-220-VHHD-2
4.75
BSC SQ
0.20 REF
0.05 MAX
0.02 NOM
0.60 MAX
0.50
BSC
0.50
0.40
0.30
COPLANARIT Y
Figure 32. 32-Lead Lead Frame Chip Scale Package [LFCSP_VQ]
5 mm × 5 mm Body, Very Thin Quad
(CP-32-2)
Dimensions shown in millimeters
ORDERING GUIDE
Model Temperature Range Package Description Package Option