SNR: 75.2 dBFS in a 55 MHz band to 185 MHz at 250 MSPS
SNR: 72.8 dBFS in an 82 MHz band to 185 MHz at 250 MSPS
Performance with NSR disabled
SNR: 66.4 dBFS up to 185 MHz at 250 MSPS
SFDR: 87 dBc up to 185 MHz at 250 MSPS
Total power consumption: 358 mW at 250 MSPS
1.8 V supply voltages
LVDS (ANSI-644 levels) outputs
Integer 1-to-8 input clock divider (625 MHz maximum input)
Internal ADC voltage reference
Flexible analog input range
1.4 V p-p to 2.0 V p-p (1.75 V p-p nominal)
Differential analog inputs with 350 MHz bandwidth
Serial port control
Energy saving power-down modes
User-configurable, built-in self test (BIST) capability
APPLICATIONS
Communications
Diversity radio and smart antenna (MIMO) systems
Multimode digital receivers (3G)
WCDMA, LTE, CDMA2000
WiMAX, TD-SCDMA
I/Q demodulation systems
General-purpose software radios
VIN+
VIN–
VCM
FUNCTIONAL BLOCK DIAGRAM
VDD
PIPELINE
ADC
REFERENCE
SERIAL PORT
SCLKSDIOCSBCLK+ CLK–
AGNDDRVDD
NOISE SHAPING
REQUANTIZ ER
AD6672
Figure 1.
AD6672
1114
AND LVDS DRIVE RS
DATA MULITIPLEXER
1-TO-8
CLOCK
DIVIDER
DCO±
0/D0±
D9±/D10±
OR±
09997-001
GENERAL DESCRIPTION
The AD6672 is an 11-bit intermediate receiver with sampling
speeds of up to 250 MSPS. The AD6672 is designed to support
communications applications, where low cost, small size, wide
bandwidth, and versatility are desired.
The ADC core features a multistage, differential pipelined
architecture with integrated output error correction logic. The
ADC features wide bandwidth inputs supporting a variety of
user-selectable input ranges. An integrated voltage reference
eases design considerations. A duty cycle stabilizer is provided
to compensate for variations in the ADC clock duty cycle,
allowing the converters to maintain excellent performance.
Rev. 0
Information furnished by Analog Devices is believed to be accurate and reliable. However, no
responsibility is assumed by Anal og Devices for its use, nor for any infringements of patents or ot her
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.
The ADC core output is connected internally to a noise shaping
requantizer (NSR) block. The device supports two output modes
that are selectable via the serial port interface (SPI). With the
NSR feature enabled, the outputs of the ADCs are processed such
that the AD6672 supports enhanced SNR performance within a
limited region of the Nyquist bandwidth while maintaining an
11-bit output resolution. The NSR block is programmed to provide
a bandwidth of up to 33% of the sample clock. For example, with
a sample clock rate of 250 MSPS, the AD6672 can achieve up to
73.6 dBFS SNR for an 82 MHz bandwidth at 185 MHz f
With the NSR block disabled, the ADC data is provided directly
to the output with an output resolution of 11 bits. The AD6672
can achieve up to 66.6 dBFS SNR for the entire Nyquist bandwidth
when operated in this mode.
When the NSR block is disabled, the ADC data is provided directly
to the output at a resolution of 11 bits. This allows the AD6672
to be used in telecommunication applications, such as a digital
predistortion observation path, where wider bandwidths are
required.
After digital signal processing, multiplexed output data is
routed into one 11-bit output port such that the maximum
data rate is 500 Mbps (DDR). This output is LVDS and
supports ANSI-644 levels.
The AD6672 receiver digitizes a wide spectrum of IF frequencies.
This IF sampling architecture greatly reduces component cost
and complexity compared with traditional analog techniques or
less integrated digital methods.
Flexible power-down options allow significant power savings.
Programming for device setup and control is accomplished
using a 3-wire, SPI-compatible serial interface with numerous
modes to support board level system testing.
The AD6672 is available in a 32-lead, RoHS-compliant LFCSP
and is specified over the industrial temperature range of −40°C
to +85°C. This product is protected by a U.S. patent.
PRODUCT HIGHLIGHTS
1. Integrated 11-bit, 250 MSPS ADC with a noise shaping
requantizer option.
2. Operation from a single 1.8 V supply and a separate digital
output driver supply accommodating LVDS outputs.
3. On-chip 1-to-8 integer clock divider function to support a
wide range of clocking.
4. Noise shaping requantizer function allows attaining improved
SNR within a reduced frequency band. With NSR enabled,
the AD6672 supports up to 82 MHz at 250 MSPS.
5. Standard serial port interface (SPI) that supports various
product features and functions, such as data formatting
(offset binary, twos complement, or gray coding), enabling
the clock DCS, power-down, test modes, and voltage
reference mode.
No Missing Codes Full Guaranteed
Offset Error Full ±11 mV
Gain Error Full +3/−6.5 % FSR
Differential Nonlinearity (DNL) Full ±0.2 LSB
25°C ±0.1 LSB
Integral Nonlinearity (INL)1 Full ±0.3 LSB
25°C ±0.12 LSB
TEMPERATURE DRIFT
Offset Error Full ±7 ppm/°C
Gain Error Full ±85 ppm/°C
INPUT-REFERRED NOISE
VREF = 1.0 V 25°C 0.65 LSB rms
ANALOG INPUT
Input Span Full 1.75 V p-p
Input Capacitance2 Full 5 pF
Input Resistance Full 20 kΩ
Input Common-Mode Voltage Full 0.9 V
POWER SUPPLIES
Supply Voltage
AVDD Full 1.7 1.8 1.9 V
DRVDD Full 1.7 1.8 1.9 V
Supply Current
1
I
Full 136 145 mA
AVDD
1
I
(NSR Disabled) Full 63 68 mA
DRVDD
1
I
(NSR Enabled, 22% Bandwidth Mode) Full 89 mA
DRVDD
1
I
(NSR Enabled, 33% Bandwidth Mode) Full 99 mA
DRVDD
POWER CONSUMPTION
Sine Wave Input (DRVDD = 1.8 V, NSR Disabled) Full 358 385 mW
Sine Wave Input (DRVDD = 1.8 V, NSR Enabled,
22% Bandwidth Mode) Full 405
Sine Wave Input (DRVDD = 1.8 V, NSR Enabled,
33% Bandwidth Mode) Full 423
Standby Power3 Full 50 mW
Power-Down Power Full 5 mW
1
Measured with a low input frequency, full-scale sine wave, with approximately 5 pF loading on each output bit.
2
Input capacitance refers to the effective capacitance between one differential input pin and AGND. See for the equivalent analog input structure. Figure 18
3
Standby power is measured with a dc input, the CLK pin inactive (set to AVDD or AGND).
Logic Compliance CMOS/LVDS/LVPECL
Internal Common-Mode Bias Full 0.9 V
Differential Input Voltage Full 0.3 3.6 V p-p
Input Voltage Range Full AGND AVDD V
Input Common-Mode Range Full 0.9 1.4 V
High Level Input Current Full 10 +22 µA
Low Level Input Current Full −22 −10 µA
Input Capacitance Full
4
Input Resistance Full 12 15 18 kΩ
LOGIC INPUT (CSB)1
High Level Input Voltage Full 1.22 2.1 V
Low Level Input Voltage Full 0 0.6 V
High Level Input Current Full 50 71 µA
Low Level Input Current Full −5 +5 µA
Input Resistance Full 26 kΩ
Input Capacitance Full 2 pF
LOGIC INPUT (SCLK)2
High Level Input Voltage Full 1.22 2.1 V
Low Level Input Voltage Full 0 0.6 V
High Level Input Current Full 45 70 µA
Low Level Input Current Full −5 +5 µA
Input Resistance Full 26 kΩ
Input Capacitance Full 2 pF
LOGIC INPUTS (SDIO)1
High Level Input Voltage Full 1.22 2.1 V
Low Level Input Voltage Full 0 0.6 V
High Level Input Current Full 45 70 µA
Low Level Input Current Full −5 +5 µA
Input Resistance Full 26 kΩ
Input Capacitance Full 5 pF
DIGITAL OUTPUTS (OR+, OR−)
LVDS Data and OR Outputs
Differential Output Voltage (VOD), ANSI Mode Full 250 350 450 mV
Output Offset Voltage (VOS), ANSI Mode Full 1.15 1.25 1.35 V
Differential Output Voltage (VOD), Reduced Swing Mode Full 150 200 280 mV
Output Offset Voltage (VOS), Reduced Swing Mode Full 1.15 1.25 1.35 V
1
Pull-up.
2
Pull-down.
pF
Rev. 0 | Page 7 of 32
Page 8
AD6672
SWITCHING SPECIFICATIONS
Table 4.
Parameter Temperature Min Typ Max Unit
CLOCK INPUT PARAMETERS
Input Clock Rate Full 625 MHz
Conversion Rate1 Full 40 250 MSPS
CLK Period—Divide-by-1 Mode (t
CLK Pulse Width High (tCH)
Divide-by-1 Mode, DCS Enabled Full 1.8 2.0 2.2 ns
Divide-by-1 Mode, DCS Disabled Full 1.9 2.0 2.1 ns
Divide-by-2 Mode Through Divide-by-8 Mode Full 0.8 ns
Aperture Delay (tA) Full 1.0 ns
Aperture Uncertainty (Jitter, tJ) Full 0.1 ps rms
DATA OUTPUT PARAMETERS
Data Propagation Delay (tPD) Full 4.1 4.7 5.2 ns
DCO Propagation Delay (t
DCO-to-Data Skew (t
DCO
) Full 0.3 0.5 0.7 ns
SKEW
Pipeline Delay (Latency)—NSR Disabled Full 10 Cycles
Pipeline Delay (Latency)—NSR Enabled Full 13 Cycles
Wake-Up Time (from Standby) Full 10 µs
Wake-Up Time (from Power-Down) Full 100 µs
Out-of-Range Recovery Time Full 3 Cycles
1
Conversion rate is the clock rate after the divider.
Timing Diagram
VIN
) Full 4 ns
CLK
) Full 4.7 5.3 5.8 ns
t
N – 1
A
N
N + 3
N + 4
N + 5
CLK+
CLK–
DCO–
DCO+
ODD/EVEN
0/D0±
(LSB)
D9±/D10±
(MSB)
N + 1
t
CH
t
CLK
t
DCO
t
PD
N – 10D0N – 10
D9
N – 10
t
SKEW
0
N – 9D0N – 90N – 8D0N – 80N – 7
D10
N – 10D9N – 9
0
N + 2
D10
N – 9D9N – 8
D10
N – 8D9N – 7
D0
N – 7
D10
N – 7D9N – 6
0
N – 6
09997-002
Figure 2. LVDS Data Output Timing
Rev. 0 | Page 8 of 32
Page 9
AD6672
TIMING SPECIFICATIONS
Table 5.
Parameter Test Conditions/Comments Min Typ Max Unit
SPI TIMING REQUIREMENTS See Figure 42 for the SPI timing diagram
tDS Setup time between the data and the rising edge of SCLK 2 ns
tDH Hold time between the data and the rising edge of SCLK 2 ns
t
Period of the SCLK 40 ns
CLK
tS Setup time between CSB and SCLK 2 ns
tH Hold time between CSB and SCLK 2 ns
t
Minimum period that SCLK should be in a logic high state 10 ns
HIGH
t
Minimum period that SCLK should be in a logic low state 10 ns
LOW
t
EN_SDIO
t
DIS_SDIO
Time required for the SDIO pin to switch from an input to an output
relative to the SCLK falling edge (not shown in Figure 42)
Time required for the SDIO pin to switch from an output to an input
relative to the SCLK rising edge (not shown in Figure 42)
10 ns
10 ns
Rev. 0 | Page 9 of 32
Page 10
AD6672
ABSOLUTE MAXIMUM RATINGS
Table 6.
Parameter Rating
Electrical
AVDD to AGND −0.3 V to +2.0 V
DRVDD to AGND −0.3 V to +2.0 V
VIN+, VIN− to AGND −0.3 V to AVDD + 0.2 V
CLK+, CLK− to AGND −0.3 V to AVDD + 0.2 V
VCM to AGND −0.3 V to AVDD + 0.2 V
CSB to AGND −0.3 V to DRVDD + 0.3 V
SCLK to AGND −0.3 V to DRVDD + 0.3 V
SDIO to AGND −0.3 V to DRVDD + 0.3 V
0/D0−, 0/D0 + Through D9−/D10−,
−0.3 V to DRVDD + 0.3 V
D9+/D10+ to AGND
OR+/OR− to AGND −0.3 V to DRVDD + 0.3 V
DCO+, DCO− to AGND −0.3 V to DRVDD + 0.3 V
Environmental
Operating Temperature Range
−40°C to +85°C
(Ambient)
Maximum Junction Temperature
150°C
Under Bias
Storage Temperature Range
−65°C to +125°C
(Ambient)
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 CHARACTERISTICS
The exposed paddle must be soldered to the ground plane for the
LFCSP package. Soldering the exposed paddle to the customer
board increases the reliability of the solder joints, maximizing
the thermal capability of the package.
Table 7. Thermal Resistance
Airflow
Package
Typ e
32-Lead LFCSP
5 mm × 5 mm
(CP-32-12)
1
Per JEDEC 51-7, plus JEDEC 25-5 2S2P test board.
2
Per JEDEC JESD51-2 (still air) or JEDEC JESD51-6 (moving air).
3
Per MIL-Std 883, Method 1012.1.
4
Per JEDEC JESD51-8 (still air).
Veloc ity
(m/sec) θ
1, 2
JA
1, 3
θ
JC
1, 4
θ
Unit
JB
0 37.1 3.1 20.7 °C/W
1.0 32.4 °C/W
2.0 29.1 °C/W
Typical θJA is specified for a 4-layer PCB with a solid ground
plane. As shown in Ta b le 7 , airflow increases heat dissipation,
which reduces θ
. In addition, metal in direct contact with the
JA
package leads from metal traces—through holes, ground, and
power planes—reduces the θ
.
JA
ESD CAUTION
Rev. 0 | Page 10 of 32
Page 11
AD6672
2
PIN CONFIGURATIONS AND FUNCTION DESCRIPTIONS
AVD D
AVD D
VIN+
VIN–
AVD D
AVD D
VCM
D5–/D6–
D5+/D6+
D7–/D8–
DNC
25
24 CSB
23
22
21
20
19
18
17 DRVDD
D7+/D8+
SCLK
SDIO
DCO+
DCO–
D9+/D10+ (MSB)
D9–/D10– (MSB)
09997-003
32313029282726
CLK+
1
2
CLK–
AVD D
OR–
OR+
0/D0– (LSB)
0/D0+ (LS B)
DRVDD
NOTES
1. THE EXPO SED THERMAL PADDLE ON THE BOTTOM OF THE
PACKAGE PROVIDES THE ANALOG GROUND FO R THE
PART. THIS EXPOSED PADDLE MUST BE CO NNECTED TO
GROUND FOR PRO PER OPERATION.
. DNC = NO NOT CO NNECT. DO NOT CONNECT TO THI S PIN.
3
INTERLEAVED
4
5
6
7
8
9
D1–/D2–
AD6672
LVD S
TOP VIEW
(Not to Scale)
10111213141516
D3–/D4–
D1+/D2+
D3+/D4+
Figure 3. LFCSP Pin Configuration (Top View)
Table 8. Pin Function Descriptions
Pin No. Mnemonic Type Description
ADC Power Supplies
8, 17 DRVDD Supply Digital Output Driver Supply (1.8 V Nominal).
3, 27, 28, 31, 32 AVDD Supply Analog Power Supply (1.8 V Nominal).
0 AGND,
25 DNC Do Not Connect. Do not connect to this pin.
ADC Analog
30 VIN+ Input Differential Analog Input Pin (+).
29 VIN− Input Differential Analog Input Pin (−).
26 VCM Output Common-Mode Level Bias Output for Analog Inputs. This pin should be decoupled to
5 OR+ Output Overrange indicator—True.
4 OR− Output Overrange indicator—Complement.
7 0/D0+ (LSB) Output DDR LVDS Output Data 0—True. The output bit on the rising edge of the data clock output
6 0/D0− (LSB) Output DDR LVDS Output Data 0—Complement. The output bit on the rising edge of the data clock
10 D1+/D2+ Output DDR LVDS Output Data 1/2—True.
9 D1−/D2− Output DDR LVDS Output Data 1/2—Complement.
12 D3+/D4+ Output DDR LVDS Output Data 3/4—True.
11 D3−/D4− Output DDR LVDS Output Data 3/4—Complement.
14 D5+/D6+ Output DDR LVDS Output Data 5/6—True.
13 D5−/D6− Output DDR LVDS Output Data 5/6—Complement.
16 D7+/D8+ Output DDR LVDS Output Data 7/8—True.
15 D7−/D8− Output DDR LVDS Output Data 7/8—Complement.
19 D9+/D10+ (MSB) Output DDR LVDS Output Data 9/10—True.
18 D9−/D10− (MSB) Output DDR LVDS Output Data 9/10—Complement.
21 DCO+ Output LVDS Data Clock Output—True.
20 DCO− Output LVDS Data Clock Output—Complement.
SPI Control
23 SCLK Input SPI Serial Clock.
22 SDIO Input/output SPI Serial Data I/O.
24 CSB Input SPI Chip Select (Active Low).
Ground Analog Ground. The exposed thermal paddle on the bottom of the package provides the
Exposed Paddle
analog ground for the part. This exposed paddle must be connected to ground for proper
operation.
ground using a 0.1 F capacitor.
(DCO) from this output is always a Logic 0 (see Figure 2).
output (DCO) from this output is always a Logic 0 (see Figure 2).
Figure 10. Single-Tone SNR/SFDR vs. Input Amplitude (A
with f
= 90.1 MHz, fS = 250 MSPS
IN
)
IN
Figure 11. Single-Tone SNR/SFDR vs. Input Frequency (fIN), fS = 250 MSPS
Figure 13. Two-Tone SFDR/IMD3 vs. Input Amplitude (A
with f
= 184.12 MHz, f
IN1
= 187.12 MHz, fS = 250 MSPS
IN2
)
IN
Figure 14. Two-Tone FFT with f
= 89.12 MHz, f
IN1
= 92.12 MHz
IN2
Figure 12. Two-Tone SFDR/IMD3 vs. Input Amplitude (AIN)
with f
= 89.12 MHz, f
IN1
= 92.12 MHz, fS = 250 MSPS
IN2
Figure 15. Two-Tone FFT with f
= 184.12 MHz, f
IN1
= 187.12 MHz
IN2
Rev. 0 | Page 13 of 32
Page 14
AD6672
100
95
90
85
80
75
70
SNR/SFDR (d Bc and dBFS)
65
60
406080 100 120 140 160 180 200 220 240
SFDR (dBc)
SNR (dBFS)
SAMPLE RATE (MSPS)
Figure 16. Single-Tone SNR/SFDR vs. Sample Rate (f
) with fIN = 90 MHz
S
09997-016
1000
0.65LSB RMS
16384 TOTAL HITS
9000
8000
7000
6000
5000
4000
NUMBER OF HITS
3000
2000
1000
0
OUTPUT CODE
Figure 17. Grounded Input Histogram, f
N – 1N
= 250 MSPS
S
09997-017
Rev. 0 | Page 14 of 32
Page 15
AD6672
VINA
V
A
V
R
V
A
A
V
EQUIVALENT CIRCUITS
DD
DRVDD
Figure 18. Equivalent Analog Input Circuit
DD
AVD DAV DD
CLK+
0.9V
15kΩ15kΩ
Figure 19. Equivalent Clock Input Circuit
D
DD
V+
D
TAOUT–
V–
09997-018
V–
DATAOUT+
V+
CLK–
26kΩ
350Ω
09997-021
SDIO
Figure 21. Equivalent SDIO Circuit
SCLK
09997-019
26kΩ
350Ω
09997-022
Figure 22. Equivalent SCLK Input Circuit
DD
26kΩ
CSB
350Ω
09997-020
Figure 20. Equivalent LVDS Output Circuit
Figure 23. Equivalent CSB Input Circuit
09997-023
Rev. 0 | Page 15 of 32
Page 16
AD6672
THEORY OF OPERATION
The AD6672 can sample any fS/2 frequency segment from dc to
250 MHz using appropriate low-pass or band-pass filtering at
the ADC inputs with little loss in ADC performance.
Programming and control of the AD6672 are accomplished
using a 3-pin, SPI-compatible serial interface.
ADC ARCHITECTURE
The AD6672 architecture consists of a front-end sample-andhold circuit, followed by a pipelined switched-capacitor ADC.
The quantized outputs from each stage are combined into a
final 11-bit result in the digital correction logic. The pipelined
architecture permits the first stage to operate on a new input
sample and the remaining stages to operate on the preceding
samples. Sampling occurs on the rising edge of the clock.
Each stage of the pipeline, excluding the last, consists of a low
resolution flash ADC connected to a switched-capacitor digitalto-analog converter (DAC) and an interstage residue amplifier
(MDAC). The MDAC magnifies the difference between the
reconstructed DAC output and the flash input for the next stage
in the pipeline. One bit of redundancy is used in each stage to
facilitate digital correction of flash errors. The last stage simply
consists of a flash ADC.
The input stage of the AD6672 contains a differential sampling
circuit that can be ac- or dc-coupled in differential or singleended modes. The output staging block aligns the data, corrects
errors, and passes the data to the output buffers. The output
buffers are powered from a separate supply, allowing digital
output noise to be separated from the analog core. During
power-down, the output buffers go into a high impedance state.
The AD6672 features a noise shaping requantizer (NSR) to
allow higher than 11-bit SNR to be maintained in a subset of
the Nyquist band.
ANALOG INPUT CONSIDERATIONS
The analog input to the AD6672 is a differential switchedcapacitor circuit that has been designed to attain optimum
performance when processing a differential input signal.
The clock signal alternatively switches the input between
sample mode and hold mode (see the configuration shown in
Figure 24). When the input is switched into sample mode, the
signal source must be capable of charging the sampling
capacitors and settling within 1/2 clock cycle.
A small resistor in series with each input can help reduce the
peak transient current required from the output stage of the
driving source. A shunt capacitor can be placed across the
inputs to provide dynamic charging currents. This passive
network creates a low-pass filter at the ADC input; therefore,
the precise values are dependent on the application.
In intermediate frequency (IF) undersampling applications, the
shunt capacitors should be reduced. In combination with the
driving source impedance, the shunt capacitors limit the input
bandwidth. Refer to the AN-742 Application Note, Frequency Domain Response of Switched-Capacitor ADCs; the AN-827
Application Note, A Resonant Approach to Interfacing Amplifiers
to Switched-Capacitor ADCs; and the Analog Dialogue article,
“Transformer-Coupled Front-End for Wideband A/D Converters,”
for more information on this subject.
BIAS
VIN+
VIN–
S
C
S
C
PAR1
C
PAR1
C
PAR2
H
C
S
C
PAR2
S
Figure 24. Switched-Capacitor Input
BIAS
S
C
FB
S
C
S
FB
S
09997-024
For best dynamic performance, match the source impedances
driving VIN+ and VIN− and differentially balance the inputs.
Input Common Mode
The analog inputs of the AD6672 are not internally dc biased.
In ac-coupled applications, the user must provide this bias
externally. Setting the device so that V
= 0.5 × AVDD (or
CM
0.9 V) is recommended for optimum performance. An onboard common-mode voltage reference is included in the
design and is available from the VCM pin. Using the VCM
output to set the input common mode is recommended.
Optimum performance is achieved when the common-mode
voltage of the analog input is set by the VCM pin voltage
(typically 0.5 × AVDD). The VCM pin must be decoupled to
ground by a 0.1 µF capacitor, as described in the Applications
Information section. Place this decoupling capacitor close to the
pin to minimize the series resistance and inductance between
the part and this capacitor.
Rev. 0 | Page 16 of 32
Page 17
AD6672
V
Differential Input Configurations
Optimum performance can be achieved when driving the AD6672
in a differential input configuration. For baseband applications, the
AD8138, ADA4937-1, and ADA4930-1 differential drivers provide
excellent performance and a flexible interface to the ADC.
The output common-mode voltage of the ADA4930-1 is easily
set with the VCM pin of the AD6672 (see Figure 25), and the
driver can be configured in a Sallen-Key filter topology to
provide band-limiting of the input signal.
15pF
200Ω
76.8Ω
IN
0.1µF
90Ω
ADA4930-1
120Ω
200Ω
33Ω
33Ω
5pF
15pF
15Ω
15Ω
VIN–
VIN+
AVD D
ADC
VCM
0.1µF
Figure 25. Differential Input Configuration Using the ADA4930-1
For baseband applications where SNR is a key parameter,
differential transformer coupling is the recommended input
configuration. An example is shown in Figure 26. To bias the
analog input, connect the VCM voltage to the center tap of the
secondary winding of the transformer.
C2
R3
R2
VIN+
ADC
R2
VIN–
VCM
2V p-p
49.9Ω
R1
C1
R1
09997-025
the true SNR performance of the AD6672. For applications where
SNR is a key parameter, differential double balun coupling is
the recommended input configuration (see Figure 28). In this
configuration, the input is ac-coupled and the VCM voltage is
provided to the input through a 33 Ω resistor. This resistor
compensates for losses in the input baluns to provide a 50 Ω
impedance to the driver.
In the double balun and transformer configurations, the value
of the input capacitors and resistors is dependent on the input
frequency and source impedance. Based on these parameters,
the value of the input resistors and capacitors may need to be
adjusted or some components may need to be removed. Tabl e 9
displays recommended values to set the RC network for
different input frequency ranges. However, these values are
dependent on the input signal and bandwidth and should be
used only as a starting guide. Note that the values given in Tab le 9
are for each R1, R2, C2, and R3 component shown in Figure 26
and Figure 28.
Table 9. Example RC Network
Frequency
Range
(MHz)
R1
Series
(Ω)
C1
Differential
(pF)
R2
Series
(Ω)
C2
Shunt
(pF)
R3
Shunt
(Ω)
0 to 100 33 8.2 0 15 49.9
100 to 300 15 3.9 0 8.2 49.9
An alternative to using a transformer-coupled input at
frequencies in the second Nyquist zone is to use an amplifier
with variable gain. The AD8375 digital variable gain amplifier
(DVGA) provides good performance for driving the AD6672.
Figure 27 shows an example of the AD8375 driving the AD6672
through a band-pass antialiasing filter.
The signal characteristics must be considered when selecting
a transformer. Most RF transformers saturate at frequencies
below a few megahertz. Excessive signal power can also cause
core saturation, which leads to distortion.
At input frequencies in the second Nyquist zone and above, the
noise performance of most amplifiers is not adequate to achieve
1. ALL INDUCTORS ARE COILCRAFT
WITH THE EXCEPTION OF THE 1µH CHOKE INDUCTORS (0603LS).
2. FILTER VALUES SHOWN ARE FOR A 20MHz BANDWIDTH FILTER
CENTEREDAT 140MHz.
165Ω
5.1pF 3.9pF
165Ω
220nH
®
0603CS COMPONENTS
15pF
VCM
1nF
68nH
AD6672
2.5kΩ║2pF
09997-027
Figure 27. Differential Input Configuration Using the AD8375
C2
R3
33Ω
33Ω
0.1µF
R1
C1
R1R2R2
C2
VIN+
ADC
0.1µF
VCM
09997-028
VIN–
R3
Page 18
AD6672
A
V
VOLTAGE REFERENCE
A stable and accurate voltage reference is built into the AD6672.
The full-scale input range can be adjusted by varying the
reference voltage via SPI. The input span of the ADC tracks
reference voltage changes linearly.
CLOCK INPUT CONSIDERATIONS
For optimum performance, the AD6672 sample clock inputs,
CLK+ and CLK−, should be clocked with a differential signal.
The signal is typically ac-coupled into the CLK+ and CLK− pins
via a transformer or via capacitors. These pins are biased
internally (see Figure 29) and require no external bias. If the
inputs are floated, the CLK− pin is pulled low to prevent
spurious clocking.
DD
CLOCK
390pF390pF
INPUT
1nF
Figure 31. Balun-Coupled Differential Clock (Up to 625 MHz)
If a low jitter clock source is not available, another option is to
ac-couple a differential PECL signal to the sample clock input
pins as shown in Figure 32. The AD9510, AD9511, AD9512,
The AD6672 has a very flexible clock input structure. Clock
input can be a CMOS, LVDS, LVPECL, or sine wave signal.
Regardless of the type of signal being used, clock source jitter
is of the most concern, as described in the Jitter Considerations
section.
Figure 30 and Figure 31 show two preferable methods for
clocking the AD6672 (at clock rates of up to 625 MHz). A low
jitter clock source is converted from a single-ended signal to a
differential signal using an RF balun or RF transformer.
The RF balun configuration is recommended for clock frequencies
between 125 MHz and 625 MHz, and the RF transformer is recommended for clock frequencies from 10 MHz to 250 MHz. The
back-to-back Schottky diodes across the secondary winding of
the transformer limit clock excursions into the AD6672 to
approximately 0.8 V p-p differential. This limit helps prevent
the large voltage swings of the clock from feeding through to
other portions of the AD6672 while preserving the fast rise and
fall times of the signal, which are critical for low jitter
performance.
CLOCK
INPUT
CLOCK
INPUT
0.1µF
0.1µF
50kΩ50kΩ
AD95xx,
ADCLK9xx
PECL DRIVER
0.1µF
100Ω
0.1µF
240Ω240Ω
ADC
CLK+
CLK–
Figure 32. Differential PECL Sample Clock (Up to 625 MHz)
A third option is to ac-couple a differential LVDS signal to the
sample clock input pins, as shown in Figure 33. The AD9510,
Figure 33. Differential LVDS Sample Clock (Up to 625 MHz)
0.1µF
100Ω
0.1µF
ADC
CLK+
CLK–
Input Clock Divider
The AD6672 contains an input clock divider with the ability to
divide the input clock by integer values between 1 and 8. For
divide ratios other than 1, the duty cycle stabilizer (DCS) is
enabled by default on power-up.
9997-032
9997-033
CLOCK
INPUT
50Ω
Mini-Circuits
ADT1-1WT, 1:1Z
100Ω
XFMR
®
390pF390pF
390pF
SCHOTTKY
DIODES:
HSMS2822
CLK+
CLK–
ADC
09997-043
Figure 30. Transformer-Coupled Differential Clock (Up to 250 MHz)
Rev. 0 | Page 18 of 32
Page 19
AD6672
Clock Duty Cycle
Typical high speed ADCs use both clock edges to generate a
variety of internal timing signals and, as a result, may be
sensitive to clock duty cycle. Commonly, a ±5% tolerance is
required on the clock duty cycle to maintain dynamic
performance characteristics.
The AD6672 contains a DCS that retimes the nonsampling
(falling) edge, providing an internal clock signal with a nominal
50% duty cycle. This allows the user to provide a wide range of
clock input duty cycles without affecting the performance of the
AD6672.
Jitter on the rising edge of the input clock is still of paramount
concern and is not reduced by the duty cycle stabilizer. The
duty cycle control loop does not function for clock rates less
than 40 MHz nominally. The loop has a time constant
associated with it that must be considered when the clock rate
may change dynamically. A wait time of 1.5 µs to 5 µs is
required after a dynamic clock frequency increase or decrease
before the DCS loop is relocked to the input signal. During the
time that the loop is not locked, the DCS loop is bypassed, and
internal device timing is dependent on the duty cycle of the
input clock signal. In such applications, it may be appropriate to
disable the duty cycle stabilizer. In all other applications,
enabling the DCS circuit is recommended to maximize ac
performance.
Jitter Considerations
High speed, high resolution ADCs are sensitive to the quality of
the clock input. The degradation in SNR at a given input
frequency (f
SNR
) due to jitter (tJ) can be calculated by
IN
= −10 log[(2π × fIN × t
HF
)2 + 10]
JRMS
)10/(LFSNR−
In the equation, the rms aperture jitter represents the root-meansquare of all jitter sources, which include the clock input, the
analog input signal, and the ADC aperture jitter specification.
IF undersampling applications are particularly sensitive to jitter,
as shown in Figure 34.
80
75
70
65
SNR (dBFS)
60
55
50
110100
INPUT FREQ UENCY (MHz)
Figure 34. SNR vs. Input Frequency and Jitter
0.05ps
0.2ps
0.5ps
1ps
1.5ps
MEASURED
1k
09997-034
In cases where aperture jitter may affect the dynamic range of the
AD6672, treat the clock input as an analog signal. In addition,
use separate power supplies for the clock drivers and the ADC
output driver to avoid modulating the clock signal with digital
noise. Low jitter, crystal controlled oscillators provide the best
clock sources. If the clock is generated from another type of
source (by gating, dividing, or another method), it should be
retimed by the original clock during the last step.
Refer to the AN-501 Application Note, Aperture Uncertainty and
ADC System Performance, and the AN-756 Application Note,
Sampled Systems and the Effects of Clock Phase Noise and Jitter, for
more information about jitter performance as it relates to ADCs.
POWER DISSIPATION AND STANDBY MODE
As shown in Figure 35, the power dissipated by the AD6672 is
proportional to its sample rate. The data in Figure 35 was taken
using the same operating conditions as those used for the
Typical Performance Characteristics section.
Figure 35. AD6672-250 Power and Current vs. Sample Rate
IAVDD
IDRVDD
ENCODE FREQ UENCY (MSP S)
By setting the internal power-down mode bits (Bits[1:0]) in the
power modes register (Address 0x08) to 01, the AD6672 is
placed in power-down mode. In this state, the ADC typically
dissipates 2.5 mW. During power-down, the output drivers are
placed in a high impedance state.
Low power dissipation in power-down mode is achieved by
shutting down the reference, reference buffer, biasing networks,
and clock. Internal capacitors are discharged when entering
power-down mode and then must be recharged when returning
to normal operation. As a result, the wake-up time is related to
the time spent in power-down mode, and shorter power-down
cycles result in proportionally shorter wake-up times.
When using the SPI port interface, the user can place the ADC
in power-down mode or standby mode. Standby mode allows
the user to keep the internal reference circuitry powered when
faster wake-up times are required. To put the part into standby
mode, set the internal power-down mode bits (Bits[1:0]) in the
power modes register (Address 0x08) to 10. See the Memory
Map section and the AN-877 Application Note, Interfacingto High Speed ADCs via SPI, for additional details.
0.25
0.20
0.15
0.10
SUPPLY CURRENT (A)
0.05
0
09997-035
Rev. 0 | Page 19 of 32
Page 20
AD6672
DIGITAL OUTPUTS
The AD6672 output drivers can be configured for either ANSI
LVDS or reduced swing LVDS using a 1.8 V DRVDD supply.
As detailed in the AN-877 Application Note, Interfacingto High Speed ADCs via SPI, the data format can be selected for offset
binary, twos complement, or gray code when using the SPI
control.
Digital Output Enable Function (OEB)
The AD6672 has a flexible three-state ability for the digital
output pins. The three-state mode is enabled using the SPI
interface. The data outputs can be three-stated by using the
output enable bar bit (Bit 4) in Register 0x14. This OEB
function is not intended for rapid access to the data bus.
Timing
The AD6672 provides latched data with a pipeline delay of
10 input sample clock cycles when NSR is disabled and provides
13 input sample clock cycles when NSR is enabled. Data outputs
Input Span = 1.75 V p-p (V) Offset Binary Output Mode Twos Complement Mode (Default) OR
are available one propagation delay (t
the clock signal.
Minimize the length of the output data lines as well as the loads
placed on these lines to reduce transients within the AD6672.
These transients may degrade converter dynamic performance.
The lowest typical conversion rate of the AD6672 is 40 MSPS. At
clock rates below 40 MSPS, dynamic performance may degrade.
Data Clock Output (DCO)
The AD6672 also provides the data clock output (DCO)
intended for capturing the data in an external register. Figure 2
shows a timing diagram of the AD6672 output modes.
ADC OVERRANGE (OR)
The ADC overrange indicator is asserted when an overrange is
detected on the input of the ADC. The overrange condition is
determined at the output of the ADC pipeline and, therefore, is
subject to a latency of 10 ADC clock cycles. An overrange at the
input is indicated by this bit 10 clock cycles after it occurs.
) after the rising edge of
PD
Rev. 0 | Page 20 of 32
Page 21
AD6672
NOISE SHAPING REQUANTIZER
The AD6672 features a noise shaping requantizer (NSR) to
allow more than an 11-bit SNR to be maintained in a subset of
the Nyquist band. The harmonic performance of the receiver
is unaffected by the NSR feature. When enabled, the NSR
contributes an additional 0.6 dB of loss to the input signal, such
that a 0 dBFS input is reduced to −0.6 dBFS at the output pins.
Two bandwidth (BW) modes are provided; the mode can be
selected from the SPI port. In each mode, the center frequency
of the band can be tuned such that IFs can be placed anywhere
in the Nyquist band.
22% BW NSR MODE (55 MHz BW AT 250 MSPS)
The first bandwidth mode offers excellent noise performance
over 22% of the ADC sample rate (44% of the Nyquist band)
and can be centered by setting the NSR mode bits (Bits[3:1]) in
the NSR control register (Address 0x3C) to 000. In this mode,
the useful frequency range can be set using the 6-bit tuning
word (Bits[5:0]) in the NSR tuning register (Address 0x3E).
There are 57 possible tuning words (TW); each step is 0.5% of
the ADC sample rate. The following equations describe the left
band edge (f
edge (f
f
f
f
Figure 36 to Figure 38 show the typical spectrum that can be
expected from the AD6672 in the 22% bandwidth mode for
three tuning words.
AMPLITUDE (dBFS)
), the channel center (f
0
), respectively:
1
= f
× 0.005 × TW
0
ADC
= f0 + 0.11 × f
CENTER
= f0 + 0.22 × f
1
0
250MSPS
180.1MHz @ –1.6BF S
–20
SNR = 73.4dB (75.0d BFS)
SFDR = 92dBc (IN BAND)
–40
–60
–80
–100
–120
ADC
), and the right band
CENTER
ADC
0
250MSPS
180.1MHz @ –1.6BF S
–20
SNR = 73.4dB (75.0d BFS)
SFDR = 93dBc (IN BAND)
–40
–60
–80
AMPLITUDE (dBFS)
–100
–120
–140
0255075100125
FREQUENCY (MHz )
09997-037
Figure 37. 22% Bandwidth Mode, Tuning Word = 28
0
250MSPS
180.1MHz @ –1.6BF S
–20
SNR = 73.3dB (74.9d BFS)
SFDR = 92dBc (IN BAND)
–40
–60
–80
AMPLITUDE (dBFS)
–100
–120
–140
0255075100125
FREQUENCY (MHz )
09997-038
Figure 38. 22% Bandwidth Mode, Tuning Word = 41
33% BW NSR MODE (>82 MHZ BW AT 250 MSPS)
The 33% bandwidth NSR mode offers excellent noise performance
over 33% of the ADC sample rate (66% of the Nyquist band).
The fundamental can be tuned using a low-pass, band-pass, or
high-pass filter by setting the NSR tuning word bits (Bits[5:0])
in Register 0x3E.
Figure 39 to Figure 41 show the typical spectrum that can be
expected from the AD6672 with the 33% bandwidth NSR mode
enabled for three filter settings.
–140
0255075100125
FREQUENCY (MHz )
Figure 36. 22% Bandwidth Mode, Tuning Word = 13
09997-036
Rev. 0 | Page 21 of 32
Page 22
AD6672
0
250MSPS
180.1MHz @ –1.6BF S
–20
SNR = 71.1dB (72.7d BFS)
SFDR = 92dBc (IN BAND)
–40
–60
–80
AMPLITUDE (dBFS)
–100
–120
–140
0255075100125
FREQUENCY (MHz )
Figure 39. 33% Bandwidth Mode, Tuning Word = 5
0
250MSPS
180.1MHz @ –1.6BF S
–20
SNR = 71.2dB (72.8d BFS)
SFDR = 92dBc (IN BAND)
–40
–60
–80
AMPLITUDE (dBFS)
–100
0
250MSPS
180.1MHz @ –1.6BF S
–20
SNR = 70.9dB (72.5d BFS)
SFDR = 92dBc (IN BAND)
–40
–60
–80
AMPLITUDE (dBFS)
–100
–120
–140
0255075100125
09997-039
FREQUENCY (MHz )
09997-041
Figure 41. 33% Bandwidth Mode, Tuning Word = 27
–120
–140
0255075100125
FREQUENCY (MHz )
09997-040
Figure 40. 33% Bandwidth Mode, Tuning Word = 17
Rev. 0 | Page 22 of 32
Page 23
AD6672
SERIAL PORT INTERFACE (SPI)
The AD6672 serial port interface (SPI) allows the user to
configure the converter for specific functions or operations
through a structured register space provided inside the ADC.
The SPI offers added flexibility and customization, depending
on the application. Addresses are accessed via the serial port
and can be written to or read from via the port. Memory is
organized into bytes that can be further divided into fields.
These fields are documented in the Memory Map section. For
detailed operational information, see the AN-877 Application
Note, Interfacing to High Speed ADCs via SPI.
CONFIGURATION USING THE SPI
Three pins define the SPI of this ADC: the SCLK pin, the SDIO
pin, and the CSB pin (see Tabl e 1 1 ). The SCLK (serial clock) pin
is used to synchronize the read and write data presented from
and to the ADC. The SDIO (serial data input/output) pin is a
dual-purpose pin that allows data to be sent and read from the
internal ADC memory map registers. The CSB (chip select bar)
pin is an active low control that enables or disables the read and
write cycles.
Table 11. Serial Port Interface Pins
Pin Function
SCLK
Serial clock. The serial shift clock input, which is used to
synchronize serial interface reads and writes.
SDIO
Serial data input/output. A dual-purpose pin that
typically serves as an input or an output, depending on
the instruction being sent and the relative position in the
timing frame.
CSB
Chip select bar. An active low control that gates the read
and write cycles.
The falling edge of CSB, in conjunction with the rising edge of
SCLK, determines the start of the framing. An example of the
serial timing and its definitions can be found in Figure 42 and
Tabl e 5 .
Other modes involving the CSB are available. The CSB can be
held low indefinitely, which permanently enables the device;
this is called streaming. The CSB can stall high between bytes
to allow for additional external timing. When CSB is tied high,
SPI functions are placed in a high impedance mode. This mode
turns on any SPI pin secondary functions.
During an instruction phase, a 16-bit instruction is transmitted.
Data follows the instruction phase, and its length is determined
by the W0 and W1 bits.
All data is composed of 8-bit words. The first bit of each
individual byte of serial data indicates whether a read or write
command is issued. This allows the serial data input/output
(SDIO) pin to change direction from an input to an output.
In addition to word length, the instruction phase determines
whether the serial frame is a read or write operation, allowing
the serial port to be used both to program the chip and to read
the contents of the on-chip memory. If the instruction is a
readback operation, performing a readback causes the serial
data input/output (SDIO) pin to change direction from an input
to an output at the appropriate point in the serial frame.
Data can be sent in MSB first mode or in LSB first mode. MSB
first mode is the default on power-up and can be changed via
the SPI port configuration register. For more information about
this and other features, see the AN-877 Application Note, Interfacing to High Speed ADCs via SPI.
HARDWARE INTERFACE
The pins described in Ta b l e 11 comprise the physical interface
between the user programming device and the serial port of the
AD6672. The SCLK pin and the CSB pin function as inputs
when using the SPI interface. The SDIO pin is bidirectional,
functioning as an input during write phases and as an output
during readback.
The SPI interface is flexible enough to be controlled by either
FPGAs or microcontrollers. One method for SPI configuration
is described in detail in the AN-812 Application Note, Micro-controller-Based Serial Port Interface (SPI) Boot Circuit.
The SPI port should not be active during periods when the full
dynamic performance of the converter is required. Because the
SCLK signal, the CSB signal, and the SDIO signal are typically
asynchronous to the ADC clock, noise from these signals can
degrade converter performance. If the on-board SPI bus is used
for other devices, it may be necessary to provide buffers
between this bus and the AD6672 to prevent these signals from
transitioning at the converter inputs during critical sampling
periods.
Rev. 0 | Page 23 of 32
Page 24
AD6672
SPI ACCESSIBLE FEATURES
Tabl e 1 2 provides a brief description of the general features that
are accessible via the SPI. These features are described in detail
in the AN-877 Application Note, Interfacing to High Speed ADCs via SPI. The AD6672 part-specific features are described in the
Memory Map Register Description section.
Table 12. Features Accessible Using the SPI
Feature Name Description
Mode Allows the user to set either power-down mode or standby mode
Clock Allows the user to access the DCS via the SPI
Offset Allows the user to digitally adjust the converter offset
Test I/O Allows the user to set test modes to have known data on output bits
Output Mode Allows the user to set up outputs
Output Phase Allows the user to set the output clock polarity
Output Delay Allows the user to vary the DCO delay
VREF Allows the user to set the reference voltage
Digital Processing Allows the user to enable the synchronization features
t
HIGH
t
LOW
t
CLK
CSB
t
DS
t
S
t
DH
t
H
SCLK
SDIO
DON’T CA RE
R/WW1W0A12A11A10A9A8A7
Figure 42. Serial Port Interface Timing Diagram
D5D4D3D2D1D0
DON’T CA RE
DON’T CA REDON’T CA RE
9997-042
Rev. 0 | Page 24 of 32
Page 25
AD6672
MEMORY MAP
READING THE MEMORY MAP REGISTER TABLE
Each row in the memory map register table has eight bit locations.
The memory map is roughly divided into four sections: the chip
configuration registers (Address 0x00 to Address 0x02); the transfer
register (Address 0xFF); the ADC functions registers, including
setup, control, and test (Address 0x08 to Address 0x25); and the
digital feature control registers (Address 0x3C and Address 0x3E).
The memory map register table (Ta bl e 13 ) documents the
default hexadecimal value for each hexadecimal address shown.
The Bit 7 (MSB) column is the start of the default hexadecimal
value given. For example, Address 0x14, the output mode
register, has a hexadecimal default value of 0x01. This means
that Bit 0 = 1 and the remaining bits are 0s. This setting is the
default output format value, which is twos complement. For
more information on this function and others, see the AN-877
Application Note, Interfacing to High Speed ADCs via SPI. This
document details the functions controlled by Register 0x00 to
Register 0x25. The remaining registers, Register 0x3C and
Register 0x3E, are documented in the Memory Map Register
Description section.
Open Locations
All address and bit locations that are not included in Tab l e 1 3
are not currently supported for this device. Write 0s to unused
bits of a valid address location. Writing to these locations is
required only when part of an address location is open (for
example, Address 0x18). If the entire address location is open
(for example, Address 0x13), this address location should not be
written.
Default Values
After the AD6672 is reset, critical registers are loaded with
default values. The default values for the registers are given in
the memory map register table (Tabl e 13 ).
Logic Levels
An explanation of logic level terminology follows:
•“Bit is set” is synonymous with “bit is set to Logic 1” or
“writing Logic 1 for the bit.”
•“Clear a bit” is synonymous with “bit is set to Logic 0” or
“writing Logic 0 for the bit.”
Transfer Register Map
Address 0x08 to Address 0x20, as well as Address 0x3C and
Address 0x3E, are shadowed. Writes to these addresses do not
affect part operation until a transfer command is issued by
writing 0x01 to Address 0xFF, setting the transfer bit. This allows
these registers to be updated internally and simultaneously when
the transfer bit is set. The internal update takes place when the
transfer bit is set, and then the bit autoclears.
Rev. 0 | Page 25 of 32
Page 26
AD6672
MEMORY MAP REGISTER TABLE
All address and bit locations that are not included in Tab l e 1 3 are not currently supported for this device.
Table 13. Memory Map Registers
Addr
Register
(Hex)
Name
Chip Configuration Registers
0x00
SPI port
configuration
0x01 Chip ID
0x02 Chip grade Open Open Speed grade ID
Transfer Register
0xFF Transfer Open Open Open Open Open Open Open Transfer 0x00
ADC Functions Registers
0x08 Power modes Open Open Open Open Open Open
0x09 Global clock Open Open Open Open Open Open Open
0x0B Clock divide Open Open Input clock divider phase adjust
0x0D Test mode
0x0E BIST enable Open Open Open Open Open
Bit 7
(MSB)
0 LSB first Soft reset 1 1 Soft reset LSB first 0 0x18
User test
mode
control
0 = continuous/
repeat
pattern
1 =
single
pattern,
then 0s
Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1
8-bit chip ID[7:0]
(AD6672 = 0xA4)
(default)
Open Open Open Open
Internal power-down
mode
00 = normal operation
01 = full power-down
10 = standby
11 = reserved
Clock divide ratio
000 = divide by 1
001 = divide by 2
010 = divide by 3
011 = divide by 4
100 = divide by 5
101 = divide by 6
110 = divide by 7
111 = divide by 8
Output test mode
0000 = off (default)
0001 = midscale short
0010 = positive FS
0011 = negative FS
0100 = alternating checkerboard
0101 = PN long sequence
0110 = PN short sequence
0111 = one/zero word toggle
1000 = user test mode
1001 to 1110 = unused
1111 = ramp output
Nibbles are
mirrored so
that LSB
first mode
or MSB first
mode is set
correctly,
regardless
of shift
mode.
Speed
grade ID
used to differentiate
devices;
read only.
Synchronously
transfers
data from
the master
shift
register to
the slave.
Determines
various
generic
modes
of chip
operation.
Clock
divide
values
other than
000 automatically
cause the
duty cycle
stabilizer
to become
active.
When this
register is
set, the
test data is
placed on
the output
pins in
place of
normal
data.
Rev. 0 | Page 26 of 32
Page 27
AD6672
Default
Addr
Register
(Hex)
Name
0x10 Offset adjust Open Open
0x14 Output mode Open Open Open
0x15 Output adjust Open Open Open Open LVDS output drive current adjust
0x16
Clock phase
control
0x17
DCO output
delay
0x18
Input span
select
0x19
User Test
Pattern 1 LSB
0x1A
User Test
Pattern 1 MSB
0x1B
User Test
Pattern 2 LSB
0x1C
User Test
Pattern 2 MSB
0x1D
User Test
Pattern 3 LSB
0x1E
User Test
Pattern 3 MSB
0x1F
User Test
Pattern 4 LSB
0x20
User Test
Pattern 4 MSB
0x24
BIST signature
LSB
0x25
BIST signature
MSB
Digital Feature Control Registers
0x3C NSR control Open Open Open Open NSR mode
0x3E
NSR tuning
word
Bit 7
(MSB)
Invert
DCO
clock
Enable
DCO
clock
delay
Open Open Open Full-scale input voltage selection
Open Open NSR tuning word
Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1
Offset adjust in LSBs from +31 to −32
(twos complement format)
Output
enable bar
0 = on
1 = off
Open Open Open Open Open Open Open 0x00
Open Open DCO clock delay
User Test Pattern 1[7:0] 0x00
User Test Pattern 1[15:8] 0x00
User Test Pattern 2[7:0] 0x00
User Test Pattern 2[15:8] 0x00
User Test Pattern 3[7:0] 0x00
User Test Pattern 3[15:8] 0x00
User Test Pattern 4[7:0] 0x00
User Test Pattern 4[15:8] 0x00
BIST signature[7:0] 0x00 Read only.
BIST signature[15:8] 0x00 Read only.
equations for the tuning word are dependent on the NSR mode)
Open
0000 = 3.72 mA output drive current
0001 = 3.5 mA output drive current (default)
0010 = 3.30 mA output drive current
0011 = 2.96 mA output drive current
0100 = 2.82 mA output drive current
0101 = 2.57 mA output drive current
0110 = 2.27 mA output drive current
0111 = 2.0 mA output drive current (reduced range)
1000 to 1111 = reserved
Configures
the
outputs
and the
format of
the data.
Full-scale
input
adjustment
in 0.022 V
steps.
NSR
controls.
NSR frequency
tuning
word.
Rev. 0 | Page 27 of 32
Page 28
AD6672
MEMORY MAP REGISTER DESCRIPTION
For more information on functions controlled in Register 0x00
to Register 0x25, see the AN-877 Application Note, Interfacing to High Speed ADCs via SPI.
NSR Control (Register 0x3C)
Bits[7:4]—Reserved
Bits[3:1]—NSR Mode
Bits[3:1] determine the bandwidth mode of the NSR. When
Bits[3:1] are set to 000, the NSR is configured for 22%
bandwidth mode, which provides enhanced SNR performance
over 22% of the sample rate. When Bits[3:1] are set to 001, the
NSR is configured for 33% bandwidth mode, which provides
enhanced SNR performance over 33% of the sample rate.
Bit 0—NSR Enable
The NSR is enabled when Bit 0 is high and disabled when Bit 0
is low.
NSR Tuning Word (Register 0x3E)
Bits[7:6]—Reserved
Bits[5:0]—NSR Tuning Word
The NSR tuning word sets the band edges of the NSR band. In
22% bandwidth mode, there are 57 possible tuning words; in
33% bandwidth mode, there are 34 possible tuning words. In
either mode, each step represents 0.5% of the ADC sample rate.
For the equations that are used to calculate the tuning word
based on the bandwidth mode of operation, see the Noise
Shaping Requantizer section.
Rev. 0 | Page 28 of 32
Page 29
AD6672
APPLICATIONS INFORMATION
DESIGN GUIDELINES
Before starting system level design and layout of the AD6672,
it is recommended that the designer become familiar with these
guidelines, which discuss the special circuit connections and
layout requirements for certain pins.
Power and Ground Recommendations
When connecting power to the AD6672, it is recommended that
two separate 1.8 V supplies be used: use one supply for analog
(AVDD) and a separate supply for the digital outputs (DRVDD).
The designer can employ several different decoupling capacitors
to cover both high and low frequencies. Locate these capacitors
close to the point of entry at the PC board level and close to the
pins of the part with minimal trace length.
A single PCB ground plane should be sufficient when using the
AD6672. With proper decoupling and smart partitioning of the
PCB analog, digital, and clock sections, optimum performance
can be easily achieved.
Exposed Paddle Thermal Heat Slug Recommendations
It is mandatory that the exposed paddle on the underside of the
ADC be connected to analog ground (AGND) to achieve the
best electrical and thermal performance. A continuous, exposed
(no solder mask) copper plane on the PCB should mate to the
AD6672 exposed paddle, Pin 0.
The copper plane should have several vias to achieve the lowest
possible resistive thermal path for heat dissipation to flow
through the bottom of the PCB. These vias should be filled or
plugged with nonconductive epoxy.
To maximize the coverage and adhesion between the ADC
and the PCB, overlay a silkscreen to partition the continuous
plane on the PCB into several uniform sections. This provides
several tie points between the ADC and the PCB during the
reflow process. Using one continuous plane with no partitions
guarantees only one tie point between the ADC and the PCB.
See the evaluation board for a PCB layout example. For detailed
information about the packaging and PCB layout of chip scale
packages, refer to the AN-772 Application Note, A Design and
Manufacturing Guide for the Lead Frame Chip Scale Package
(LFCSP).
VCM
Decouple the VCM pin to ground with a 0.1 F capacitor, as
shown in Figure 26.
SPI Port
The SPI port should not be active during periods when the full
dynamic performance of the converter is required. Because the
SCLK, CSB, and SDIO signals are typically asynchronous to the
ADC clock, noise from these signals can degrade converter
performance. If the on-board SPI bus is used for other devices,
it may be necessary to provide buffers between this bus and the
AD6672 to keep these signals from transitioning at the
converter input pins during critical sampling periods.
Rev. 0 | Page 29 of 32
Page 30
AD6672
OUTLINE DIMENSIONS
PIN 1
INDICATOR
5.10
5.00 SQ
4.90
0.50
BSC
24
0.30
0.25
0.18
25
EXPOSED
PAD
P
N
I
32
1
N
I
*
3.75
3.60 SQ
3.55
1
A
O
R
T
D
C
I
0.80
0.75
0.70
SEATING
PLANE
0.50
0.40
0.30
0.05 MAX
0.02 NOM
COPLANARITY
0.20 REF
*
COMPLIANT TO JEDEC STANDARDS MO-220-WHHD-5
WITH EXCEPTION TO EXPOSED PAD DIMENSION.
17
BOTTOM VIEWTOP VIEW
0.08
8
916
0.25 MIN
FOR PROPER CONNECTION OF
THE EXPOSED PAD, REFER TO
THE PIN CONFIGURATION AND
FUNCTION DESCRIPTIONS
SECTION OF THIS DATA SHEET.
08-16-2010-B
Figure 43. 32-Lead Lead Frame Chip Scale Package [LFCSP_WQ]
5 mm × 5 mm Body, Very Very Thin Quad
(CP-32-12)
Dimensions shown in millimeters
ORDERING GUIDE
Model1 Temperature Range Package Description Package Option
AD6672BCPZ-250 −40°C to +85°C 32-Lead Lead Frame Chip Scale Package [LFCSP_WQ] CP-32-12
AD6672BCPZRL7-250 −40°C to +85°C 32-Lead Lead Frame Chip Scale Package [LFCSP_WQ] CP-32-12
AD6672-250EBZ −40°C to +85°C Evaluation Board with AD6672 and Software