Serial Interface, 625 kSPS, 24-Bit
Σ∆
ADC
Preliminary Technical Data
AD7763
Rev. PrD | Page 1 of 24
FEATURES
High performance 24-bit Sigma-Delta ADC
118dB SNR at 78kHz output data rate
106dB SNR at 625 kHz output data rate
625 kHz maximum fully filtered output word rate
Programmable over-sampling rate (32x to 256x)
Flexible serial interface
Fully differential modulator input
On-chip differential amplifier for signal buffering
Low pass FIR filter with default or user programmable
coefficients
Over-range alert bit
Digital offset and gain correction registers
Low power and power down modes
Synchronization of multiple devices via SYNC pin
APPLICATIONS
Data acquisition systems
Vibration analysis
Instrumentation
FUNCTIONAL BLOCK DIAGRAM
DIFF
Multi-Bit
Sigma-Delta
Modulator
AD7763
Control Logic,
I/O and
Registers
Reconstruction
Programmable
Decimation
FIR Filter
Engine
VIN+VIN-
AVDD1
AGND
VREF+
MCLK
SDEN
DGND
BUF
VDRIVE
AVDD2
AVDD3
AVDD4
DVDD
-
+
DECAP1
RBIAS
MCLKGND
ADR2:0
CLKDIV
DECAP2
DECAP3
REFGND
SCP
I2S
SCR
SDL
DRDY
SCO
SDO
SDI
FSI
FSO
RESET
SYNC
SH2:0
Figure 1.
PRODUCT OVERVIEW
The AD7763 high performance, 24-bit, sigma delta analog to
digital converter combines wide input bandwidth and high
speed with the benefits of sigma delta conversion, as well as
performance of 106dB SNR at 625 kSPS making it ideal for high
speed data acquisition. A wide dynamic range combined with
significantly reduced anti-aliasing requirements simplifies the
design process. An integrated buffer to drive the reference, a
differential amplifier for signal buffering and level shifting, an
over-range flag, internal gain & offset registers and a low-pass
digital FIR filter make the AD7763 a compact highly integrated
data acquisition device requiring minimal peripheral
component selection. In addition the device offers
programmable decimation rates and the digital FIR filter can be
adjusted if the default characteristics are not appropriate to the
application. The AD7763 is ideal for applications demanding
high SNR without necessitating design of complex front end
signal processing.
The differential input is sampled at up to 40MS/s by an analog
modulator. The modulator output is processed by a series of
low-pass filters, the final one having default or user
programmable coefficients. The sample rate, filter corner
frequencies and output word rate are set by a combination of
the external clock frequency and the configuration registers of
the AD7763.
The reference voltage supplied to the AD7763 determines the
analog input range. With a 4V reference, the analog input range
is ±3.2V differential biased around a common mode of 2V. This
common mode biasing can be achieved using the on-chip
differential amplifiers, further reducing the external signal
conditioning requirements.
The AD7763 is available in an exposed paddle 64-lead TQFP
and 48-lead CSP packages and is specified over the industrial
temperature range from -40°C to +85°C.
Rev. PrD
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 www.analog.com
Fax: 781.326.8703 © 2005 Analog Devices, Inc. All rights reserved.
AD7763 Preliminary Technical Data
Rev. PrD | Page 2 of 24
TABLE OF CONTENTS
TABLE OF CONTENTS.................................................................. 2
AD7763—Specifications.................................................................. 3
Timing Specifications....................................................................... 6
Timing Diagrams.............................................................................. 7
Absolute Maximum Ratings............................................................ 8
ESD Caution.................................................................................. 8
Pin Configuration and Functional Descriptions.......................... 9
Terminology .................................................................................... 11
Typical Performance Characteristics ........................................... 12
Theory of Operation ...................................................................... 13
AD7763 Interface............................................................................ 14
Clocking the AD7763 .................................................................... 16
Driving The AD7763 ..................................................................... 17
Using The AD7763..................................................................... 17
Bias Resistor Selection ............................................................... 18
Programmable FIR Filter............................................................... 19
Downloading a User-Defined Filter ............................................ 20
Example Filter Download ......................................................... 20
AD7763 Registers........................................................................... 22
Non Bit-Mapped Registers........................................................ 23
Outline Dimensions....................................................................... 24
Ordering Guide .......................................................................... 24
REVISION HISTORY
Rev PrA: Preliminary Datasheet
Rev PrB: Preliminary Datasheet
Rev PrC: Preliminary Datasheet
Preliminary Technical Data AD7763
Rev. PrD | Page 3 of 24
AD7763—SPECIFICATIONS
Table 1. VDD1 = 2.5 V, VDD2 = 5 V, VREF = 4.096 V, TA = +25°C, Full Power Mode using on-chip amplifierwith components as shown in
Table 8 , unless otherwise noted
Parameter Test Conditions/Comments Specifcation Unit
DYNAMIC PERFORMANCE1
Decimate by 256 MCLK = 40MHz, ODR = 78.125kHz, FIN = 1kHz Sine Wave
Dynamic Range 2 Modulator inputs shorted TBD dB min
120.8 dB typ
Signal to Noise Ratio (SNR)2 117 dB typ
Non-harmonic 120 dBFS typ Spurious Free Dynamic Range (SFDR) 2
Input Amplitude = -60dB 129 dBFS typ
Input Amplitude = -0.5dB -100 dB typ
TBD dB max Input Amplitude = -6dB
-103 dB typ
Total Harmonic Distortion (THD) 2
Input Amplitude = -60dB -66 dB typ
Decimate by 64 MCLK = 40MHz, ODR = 312.5kHz, FIN =100kHz Sine Wave
TBD dB min Dynamic Range2 Modulator inputs shorted
114 dB typ
TBD dB max Signal to Noise Ratio (SNR) 2 Modulator inputs shorted
109 dB typ
Spurious Free Dynamic Range (SFDR) 2 Non-harmonic 109 dBFS typ
Input Amplitude = -0.5dB -102 dB typ
TBD dB max
Total Harmonic Distortion (THD) 2
Input Amplitude = -6dB
-110 dB typ
Decimate by 32 MCLK = 40MHz, ODR = 625kHz, FIN = 100kHz Sine Wave
Dynamic Range2 Modulator inputs shorted 110 dB typ
Signal to Noise Ratio (SNR) 2 TBD dB min
104 dB typ
Spurious Free Dynamic Range (SFDR) 2 Non-harmonic 120 dBFS typ
Input Amplitude = -0.5dB -102 dB typ
TBD dB max
Total Harmonic Distortion (THD) 2
Input Amplitude = -6dB
TBD dB typ
DC ACCURACY
Resolution 24 Bits
Differential Nonlinearity2 Guaranteed monotonic to 24 bits 1 LSB typ
Integral Nonlinearity2 At 16 bits 1 LSB typ
Zero Error2 0.03 % typ
Gain Error2 0.01 % typ
Zero Error Drift 0.0006 % /°C typ
Gain Error Drift 1.4 ppm /°C
typ
DIGITAL FILTER RESPONSE
Decimate by 32
Group Delay MCLK = 40MHz TBD µS typ
Decimate by 64
Group Delay MCLK = 40MHz TBD µS typ
Decimate by 128
Group Delay MCLK = 24.576MHz 480 µS typ
ANALOG INPUT
Differential Input Voltage Vin(+) – Vin(-), VREF = 2.5V ±2 V pk-pk
Vin(+) – Vin(-), VREF = 4.096V ±3.25 V pk-pk
DC Leakage Current ±2 µA max
AD7763 Preliminary Technical Data
Rev. PrD | Page 4 of 24
Parameter Test Conditions/Comments Specifcation Unit
Input Capacitance At internal buffer inputs 5 pF typ
At modulator inputs 55 pF typ
REFERENCE INPUT/OUTPUT
VREF Input Voltage VDD3 = 3.3V +2.5 Volts
V
DD3 = 5V +4.096 Volts
VREF Input DC Leakage Current ±1 µA max
VREF Input Capacitance 5 pF max
POWER REQUIREMENTS
AVDD1 (Modulator Supply) ±5% +2.5 Volts
AVDD2 (General Supply) ±5% +5 Volts
AVDD3 (Diff-Amp Supply) +3.15/+5.25 V min/max
AVDD4 (Ref Buffer Supply) +3.15/+5.25 V min/max
DVDD ±5% +2.5 Volts
VDRIVE +1.65/+2.7 V min/max
Full Power Mode
AIDD1 (Modulator) 50 mA typ
AIDD2 (General) 35 mA typ
AIDD4 (Reference Buffer) AVDD4 = +5V 35 mA typ
Low Power Mode
AVDD1 (Modulator) 26 mA typ
AIDD2 (General) 20 mA typ
AIDD4 (Reference Buffer) AVDD4 = +5V 10 mA typ
AIDD3 (Diff Amp) AVDD3 = +5V, Both Modes 42 mA typ
DIDD Both Modes 45 mA typ
Standby Mode
AVDD1 (Modulator) 210 µA typ
AIDD2 (General) 30 nA typ
AIDD3 (Diff Amp AVDD3 = +5V 30 nA typ
AIDD4 (Reference Buffer) AVDD4 = +5V 30 nA typ
DIDD Clock Stopped 250 µA typ
Clock Running 690 µA typ
POWER DISSIPATION
Full Power Mode
Modulator (P1) 125 mW typ
General (P2) 175 mW typ
Reference Buffer (P4) AVDD4 = +3.3V 101 mW typ
AVDD4 = +5V 175 mW typ
Low Power Mode
Modulator (P1) 65 mW typ
General (P2) 100 mW typ
Reference Buffer (P4) AVDD4 = +3.3V 27 mW typ
AVDD4 = +5V 50 mW typ
Differential Amplifier (P3) AVDD3 = +3.3V 116 mW typ
AVDD3 = +5V 210 mW typ
Preliminary Technical Data AD7763
Rev. PrD | Page 5 of 24
Parameter Test Conditions/Comments Specifcation Unit
Digital Power 112.5 mW typ
Total Power Dissipation Full Power Mode TBD mW max
Low Power Mode TBD mW max
Standby Mode Clock Stopped 1.2 mW typ
Clock Running 2.3 mW typ
DIGITAL I/O
Input Capacitance TBD pF typ
VINH TBD V min
VINL TBD V max
VOH TBD V min
VOL TBD V max
1 All specifications in dB’s are referred to a full-scale input, FS. Tested with an input signal at 0.5dB below full-scale, unless otherwise specified.
2 See Terminology
AD7763 Preliminary Technical Data
Rev. PrD | Page 6 of 24
TIMING SPECIFICATIONS
Table 2. VDD1 = 2.5 V, VDD2 = 5 V, VREF = 4.096 V, VDRIVE = TBD V, TA = +25°C, CLOAD = 25pF, Full Power Mode, unless otherwise noted
Parameter Limit at TMIN, TMAX Unit Description
fMCLK 1 MHz min Applied Master Clock Frequency
40 MHz max
fICLK 500 kHz min Internal Modulator Clock Derived from MCLK.
20 MHz max
t13 1 × tICLK or 0.5 × tICLK4 typ SCO High Period
t2 1 × tICLK or 0.5 × tICLK typ SCO Low Period
t3 TBD typ
DRDY Low Period
t4 32 × tSCO5 typ
FSO Low Period
t5 TBD ns max Initial Data Access Time
t6 TBD6 ns max SCO Rising Edge to SDO Valid
t7 TBD nS min SDO Valid after SCO Falling Edge
t8 8/16 × tSCO typ
FSO Falling Edge to SDL Falling Edge
t9 TBD typ SDL Pulse Width
t10 TBD ns max
FSO Rising Edge to SDO Three-state
t11 TBD min
FSI Low Period
t12 TBD xS min SDI Setup Time
t13 TBD xS min SDI Hold Time
3 tICLK = 1/fICLK
4 SCO frequency selected by SCR pin
5 tSCO = t1+t2
6 All edges mentioned refer to SCP=0. Invert SCO edges for SCP=1
Preliminary Technical Data AD7763
Rev. PrD | Page 7 of 24
TIMING DIAGRAMS
t1
t2
t7
t3
t4
t5t6
t8t9
D23 D22 D21 D20 D19 D18 D17 D16 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 ST7 ST6 ST5 ST4 ST3 ST2 ST1 ST0
t10
SCO(O)
DRDY (O)
FSO (O)
SDO (O)
SDL (O)
Figure 2. Serial Read Timing Diagram
t1t2
32 SCLKs
t11
t12 t13
ALL AD2 AD1 AD0 RA11 RA10 RA1 RA0 D15 D14 D1 D0
SCO (O)
FSI (I)
SDI (I)
Figure 3. AD7763 Register Write
SERIAL DATA FROM ADC0 SERIAL DATA FROM ADC1 SERIAL DATA FROM ADC2 SERIAL DATA FROM ADC3
3
2CLKs
SCO (O)
DRDY (O)
FSO 0 (O)
FSO 3 (O)
FSO 2 (O)
FSO 1 (O)
SDO (O)
Figure 4. Serial Read Timing Using Multiple AD7763 Devices
AD7763 Preliminary Technical Data
Rev. PrD | Page 8 of 24
ABSOLUTE MAXIMUM RATINGS
Table 3. TA = 25°C, unless otherwise noted.
Parameters Rating
VDD to GND TBD
VIN+ to GND TBD
VIN– to GND TBD
Digital input voltage to GND TBD
Digital output voltage to GND TBD
VREF to GND TBD
Input current to any pin except supplies7 TBD
Operating temperature range
Commercial (A, B version) −40°C to +85°C
Storage temperature range −65°C to +150°C
Junction temperature 150°C
TQFP Exposed Paddle Package
θJA thermal impedance 92.7 °C/W
θJC thermal impedance 5.1 °C/W
CSP Package
θJA thermal impedance 26.7 °C/W
θJC thermal impedance 30 °C/W
Lead temperature, soldering
Vapor phase (60 secs) 215°C
Infrared (15 secs) 220°C
ESD TBD kV
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 listed in the operational sections
of this specification is not implied. Exposure to absolute
maximum rating conditions for extended periods may affect
device reliability.
7 Transient currents of up to TBD mA do not cause SCR latch-up.
ESD CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily accumulate on the
human body and test equipment and can discharge without detection. Although this product features
proprietary ESD protection circuitry, permanent damage may occur on devices subjected to high energy
electrostatic discharges. Therefore, proper ESD precautions are recommended to avoid performance
degradation or loss of functionality.
Preliminary Technical Data AD7763
Rev. PrD | Page 9 of 24
PIN CONFIGURATION AND FUNCTIONAL DESCRIPTIONS
SDI
SDL
SCP
DECAP2
DECAP3
AGND
AGND
AGND
AVDD2
AVDD2
AGND
ADR0
ADR1
ADR2
SH0
VDRIVE
DGND
DGND
DVDD
SH1
SH2
DRDY
RESET
39
38
37
41
40
SYNC
DGND
AGND
AVDD1
36
35
34
33
42
43
44
45
46
47
48
17 18 19 20 21 22 23 24
RBIAS
AGND
VINA1+
VINA1-
VOUTA1-
VOUTA1+
AGND
AVDD3
VIN+
VIN-
AVDD2
AGND
1
2
3
4
5
6
7
8
9
10
11
12
64 63 62 61 60 59 58
VDRIVE
DGND
DGND
I2S
SCR
CDIV
SCO
SDO
DGND
PIN 1
IDENTIFIER
TOP VIEW
(Not to Scale)
AD7763
DGND
MCLK
AVDD2
AVDD1
AGND
VREF+
AGND
AVDD4
13
14
15
16
25 26 27 31302928 32
57 56 55 54 53 52 51 50 49
AGND
DECAP1
REFGND
MCLKGND
FSI
FSO
SDEN
DGND
Figure 4. 64-Lead TQFP Pin Configuration
SH0
PIN 1
IDENTIFIER
AD7763
TOP VIEW
(Not to Scale)
36
35
34
33
32
31
30
29
28
27
26
25
1
2
3
4
5
6
7
8
9
10
11
12
48
47
46
45
44
43
42
41
40
39
38
37
13
14
15
16
17
18
19
20
21
22
23
24
VDRIVE
MCLK
AVDD2
AVDD1
DECAP1
REFGND
VREF+
AVDD4
AVDD2
AVDD2
RBIAS
VINA1+
VINA1-
VOUTA1-
VOUTA1+
AVDD3
VIN+
VIN-
AVDD2
DECAP2
AGND
AGND
AVDD1
SYNC
RESET
DRDY
DVDD
VDRIVE
ADR2
ADR1
ADR0
SCP
SDL
FSI
SDI
SDO
SCO
FSO
SDEN
CDIV
SCR
I2S
DECAP3
SH1
SH2
MCLKGND
DGND
Figure 5. 48-PIN LFCSP Pin Configuration
Table 4. Pin Function Descriptions
TQFP Pin
Number
CSP Pin
Number
Pin Mnemonic Description
6, 33 5, 25 AVDD1 +2.5V power supply for modulator. These pins should be decoupled to AGND with 100nF and
10µF capacitors on each pin.
4, 14, 15,
27
4, 10, 11,
20
AVDD2 +5V power supply. These pins should be decoupled to AGND with TBD nF and TBD µF
capacitors on each pin.
24 17 AVDD3 +3.3V to +5V power supply for differential amplifier. These pins should be decoupled to AGND
with a 100nF capacitor.
12 9 AVDD4 +3.3V to +5V power supply for reference buffer. This pin should be decoupled to AGND with a
10nF capacitor in series with a 10 resistor.
5, 7, 11,
13, 16, 18,
23, 28, 31,
32, 34
23, 24,
Paddle
AGND Power supply ground for analog circuitry. In the Chip Scale package, most of the internal
AGND pads are down-bonded to the exposed paddle. This paddle then become the main
analog ground connection for the AD7763.
9 7 REFGND Reference Ground. Ground connection for the reference voltage.
41 31 DVDD +2.5V power supply for digital circuitry and FIR filter. This pin should be decoupled to DGND
with a 100nF capacitor.
44, 63 1, 32 VDRIVE Logic power supply input, +1.8V to +2.5V. The voltage supplied at these pins will determine
the operating voltage of the logic interface. Both these pins must be connected together and
tied to the same supply. Each pin should also be decoupled to DGND with a 100nF capacitor.
1, 35, 42,
43, 53, 58,
62, 64
45,
Paddle
DGND Ground Reference for digital circuitry. In the Chip Scale package, all the internal DGND pads
are down-bonded to the exposed paddle. This paddle then becomes the single ground
connection for the AD7763.
AD7763 Preliminary Technical Data
Rev. PrD | Page 10 of 24
TQFP Pin
Number
CSP Pin
Number
Pin Mnemonic Description
19 13 VINA1+ Positive Input to Differential Amplifier.
20 14 VINA1- Negative Input to Differential Amplifier.
21 15 VOUTA1- Negative Output from Differential Amplifier.
22 16 VOUTA1+ Positive Output from Differential Amplifier.
25 18 VIN+ Positive Input to the Modulator.
26 19 VIN- Negative Input to the Modulator.
10 8 VREF+ Reference Input. The input range of this pin is determined by the reference buffer supply
voltage (AVDD4). See Reference Section for more details.
8 6 DECAP1 Decoupling Pin. A 100nF capacitor must be inserted between this pin and AGND.
29 21 DECAP2 Decoupling Pin. A TBD µF capacitor must be inserted between this pin and AGND.
30 22 DECAP3 Decoupling Pin. A TBD µF capacitor must be inserted between this pin and AGND.
17 12 RBIAS Bias Current setting pin. A resistor must be inserted between this pin and AGND. For more
details on this, see the Bias Resistor Section.
37 27
RESET A falling edge on this pin resets all internal digital circuitry. Holding this pin low keeps the
AD7763 in a reset state.
3 3 MCLK Master Clock Input. A low jitter digital clock must be applied to this pin. The output data rate
will depend on the frequency of this clock. See Clocking Section for more details.
2 2 MCLKGND Master Clock ground sensing pin.
36 26
SYNC Synchronization Input. A falling edge on this pin resets the internal filter. This can be used to
synchronize multiple devices in a system.
38 28
DRDY Data Ready Output. Each time that new conversion data is available, an active low pulse, ½ICLK
period wide, is produced on this pin. See AD7763 Interface Section for further details.
39, 40, 45 29, 30,
33
SH2:0 Share Pins 2:0. For multiple AD7763 devices sharing a common serial bus, hardwiring these
pins with the number of devices sharing the bus will ensure the correct number of DRDY
pulses are generated. See Sharing the Serial Bus in the AD7763 Interface Section for further
details.
46-48 34-36 ADR2:0 Address 2:0. To allow multiple AD7763 devices share a common serial bus, each device must
be given an individual, hard-wired address. See Sharing the Serial Bus in the AD7763 Interface
Section for further details.
49 37 SCP Serial Clock Polarity. Determines which edge of SCO that the data bits are clocked out and
valid on. All timing diagrams are shown with SCP=0 and all SCO edges shown should be
inverted for SCP=1.
50 38 SDL Serial Data Latch. A pulse is output on this pin after every 16 Data bits. The pulse is one SCO
period wide and could be used in conjunction with FSO as an alternative framing method for
serial transfers requiring a framing signal more frequently than every 32 bits.
51 39
FSI Frame Sync In. The status of this pin is checked on the falling edge of SCO. If this pin is low
then the first data bit will be latched in on the next SCO falling edge.
52 40 SDI Serial Data In. The first data bit (MSB) must be valid on the next SCO falling edge after the FSI
event has been latched. 32 bits are required for each write; the first 16-bit word contains the
device and register address and the second word contains the data.
54 41 SDO Serial Data Out. Address, Status and Data bits are clocked out on this line during each serial
transfer. If SCP=0, each bit is clocked out on an SCO rising edge and valid on the falling edge.
55 42 SCO Serial Clock Out. This clock signal is derived from the internal ICLK signal. The frequency of this
clock is either equal to ICLK or ICLK/2 depending on the state of the SCR pin. See the Clocking
the AD7763 and AD7763 Interface sections for further details.
56 43
FSO Frame Sync Out. This signal frames the serial data output and is 32 SCO periods wide.
57 44
SDEN Serial Data Enable. When this pin is logic high, SDO will remain in three-state.
59 46 CLKDIV Clock Divider. This pin is used to select the ratio of MCLK to ICLK. See the Clocking the AD7763
Section for further details.
60 47 SCR Serial Clock Rate. This pin as well as the state of the CLKDIV pin programs the SCO
frequency.See Table 6 for details..
61 48 I2S I2S Select. A logic 1 on this pin changes the serial data out mode from SPI to I2S compatible.
Preliminary Technical Data AD7763
Rev. PrD | Page 11 of 24
TERMINOLOGY
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.
Total Harmonic Distortion (THD)
The ratio of the rms sum of harmonics to the fundamental. For
the AD7763, it is defined as
()
1
2
6
2
5
2
4
2
3
2
2
V
VVVVV
20dBTHD ++++
=log
where:
V1 is the rms amplitude of the fundamental.
V2, V3, V4, V5, and V6 are the rms amplitudes of the second to
the sixth harmonics.
Dynamic Range
The dynamic range is the ratio of the rms value of the full scale
to the rms noise measured with the inputs shorted together. The
value for dynamic range is expressed in decibels.
Peak Harmonic or Spurious Noise
The ratio of the rms value of the next largest component in the
ADC output spectrum (up to fS/2 and excluding dc) to the rms
value of the fundamental. Normally, the value of this
specification is determined by the largest harmonic in the
spectrum, but, for ADCs where the harmonics are buried in the
noise floor, it is a noise peak.
Non-Harmonic Spurious Free Dynamic Range (SFDR)
The ratio of the rms signal amplitude to the rms value of the
peak spurious spectral component excluding harmonics.
Intermodulation Distortion
With inputs consisting of sine waves at two frequencies, fa and
fb, any active device with nonlinearities creates distortion
products at sum and difference frequencies of mfa ± nfb, where
m, n = 0, 1, 2, 3, and so on. Intermodulation distortion terms
are those for which neither m nor n are equal to zero. For
example, the second-order terms include (fa + fb) and (fa − fb),
while the third-order terms include (2fa + fb), (2fa − fb), (fa +
2fb) and (fa − 2fb).
The AD7763 is tested using the CCIF standard, where two input
frequencies near the top end of the input bandwidth are used.
In this case, the second-order terms are usually distanced in
frequency from the original sine waves, while the third-order
terms are usually at a frequency close to the input frequencies.
As a result, the second- and third-order terms are specified
separately. The calculation of the intermodulation distortion is
as per the THD specification, where it is the ratio of the rms
sum of the individual distortion products to the rms amplitude
of the sum of the fundamentals expressed in dB.
Integral Nonlinearity (INL)
The maximum deviation from a straight line passing through
the endpoints of the ADC transfer function.
Differential Nonlinearity (DNL)
The difference between the measured and the ideal 1 LSB
change between any two adjacent codes in the ADC.
Zero Error
The zero error is the difference between the ideal midscale
input voltage (0V) and the actual voltage producing the
midscale output code,
Gain Error
The first transition (from 100…00 to 100…01) should occur for
an analog voltage ½ LSB above the nominal negative full scale.
The last transition (from 011…10 to 011…11) should occur for
an analog voltage 1 ½ LSB below the nominal full scale. 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.
AD7763 Preliminary Technical Data
Rev. PrD | Page 12 of 24
TYPICAL PERFORMANCE CHARACTERISTICS
Default Conditions: TA = 25°C, TBD, unless otherwise noted.
ALL CAPS ( Init ial cap)
000
000 000 000 000 000 000
ALL CAPS ( Init ial cap)
000
000
000
000
000
TBD
Figure 6. TBD
ALL CAPS ( Init ial cap)
000
000 000 000 000 000 000
ALL CAPS ( Init ial cap)
000
000
000
000
000
TBD
Figure 7. TBD
ALL CAPS ( Init ial cap)
000
000 000 000 000 000 000
ALL CAPS ( Init ial cap)
000
000
000
000
000
TBD
Figure 8. TBD
ALL CAPS ( Init ial cap)
000
000 000 000 000 000 000
ALL CAPS ( Init ial cap)
000
000
000
000
000
TBD
Figure 9. TBD
ALL CAPS ( Init ial cap)
000
000 000 000 000 000 000
ALL CAPS ( Init ial cap)
000
000
000
000
000
TBD
Figure 10. TBD
ALL CAPS ( Init ial cap)
000
000 000 000 000 000 000
ALL CAPS ( Init ial cap)
000
000
000
000
000
TBD
Figure 11. TBD
Preliminary Technical Data AD7763
Rev. PrD | Page 13 of 24
THEORY OF OPERATION
The AD7763 employs a sigma-delta conversion technique to
convert the analog input into an equivalent digital word. The
modulator samples the input waveform and outputs an
equivalent digital word to the digital filter at a rate equal to ICLK.
Due to the high over-sampling rate, which spreads the
quantization noise from 0 to fICLK, the noise energy contained in
the band of interest is reduced (Figure 12a). To further reduce
the quantization noise, a high order modulator is employed to
shape the noise spectrum; so that most of the noise energy is
shifted out of the band of interest (Figure 12b).
The digital filtering which follows the modulator removes the
large out-of-band quantization noise (Figure 12c) while also
reducing the data rate from fICLK at the input of the filter to fICLK/8
or less at the output of the filter, depending on the decimation
rate used.
Digital filtering has certain advantages over analog filtering. It
does not introduce significant noise or distortion and can be
made perfectly linear phase.
The AD7763 employs three Finite Impulse Response (FIR)
filters in series. By using different combinations of decimation
ratios and filter selection and bypassing, data can be obtained
from the AD7763 at a large range of data rates. The first filter
receives data from the modulator at ICLK MHz where it is
decimated by four to output data at (ICLK/4) MHz.
BAND OF INTEREST
fICLK/2
DIGITAL FILTER CUTOFF FREQUENCY
BAND OF INTEREST
QUANTIZATION NOISE
fICLK/2
BAND OF INTEREST
fICLK/2
NOISE SHAPING
a.
b.
c.
Figure 12. Sigma-Delta ADC
The second filter allows the decimation rate to be chosen from
4x to 32x or to be completely bypassed. The third filter has a
fixed decimation rate of 2x and is user programmable as well as
having a default configuration. It is described in detail in the
Programmable FIR Filter Section. This filter can also be
bypassed. Table 5 below shows some characteristics of the
default filter. The group delay of the filter is defined to be the
delay to the centre of the impulse response and is equal to the
computation + filter delays. The delay until valid data is
available (the DVALID status bit is set) is equal to 2x the filter
delay + the computation delay.
Table 5. Configuration With Default Filter
ICLK
Frequency Filter 1 Filter 2 Filter 3 Data State Computation
Delay Filter Delay Passband
Bandwidth
Output Data
Rate (ODR)
20 MHz 4x 4x 2x Fully Filtered 1.775µS 44.4µS 250 kHz 625 kHz
20 MHz 4x 8x Bypassed Partially Filtered 2.6µS 10.8µS 140.625 kHz 625 kHz
20 MHz 4x 8x 2x Fully Filtered 2.25µS 87.6µS 125 kHz 312.5 kHz
20 MHz 4x 16x Bypassed Partially Filtered 4.175µS 20.4µS 70.3125 kHz 312.5 kHz
20 MHz 4x 16x 2x Fully Filtered 3.1µS 174µS 62.5 kHz 156.25 kHz
20 MHz 4x 32x Bypassed Partially Filtered 7.325µS 39.6µS 35.156 kHz 156.25 kHz
20 MHz 4x 32x 2x Fully Filtered 4.65µS 346.8µS 31.25 kHz 78.125 kHz
12.288MHz 4x 8x 2x Fully Filtered 3.66µS 142.6µS 76.8 kHz 192 kHz
12.288MHz 4x 16x 2x Fully Filtered 5.05µS 283.2µS 38.4 kHz 96 kHz
12.288MHz 4x 32x Bypassed Partially Filtered 11.92µS 64.45µS 21.6 kHz 96 kHz
12.288MHz 4x 32x 2x Fully Filtered 7.57µS 564.5µS 19.2 kHz 48 kHz
AD7763 Preliminary Technical Data
Rev. PrD | Page 14 of 24
AD7763 INTERFACE
Reading Data
The AD7763 uses an SPI compatible serial interface. The timing
diagram in Figure 2 shows how the AD7763 transmits its
conversion results.
The data being read from the AD7763 is clocked out using the
serial clock output, SCO. The SCO frequency is dependant on
the the state of the serial clock output rate, SCR, and the clock
divider mode chosen by the state of the clock divider bit CDIV
(see the Clocking the AD7763 section). Table 6 details the SCO
frequency for the AD7763 as a result of the states of both the
CDIV and SCR pins.
Table 6. SCO Frequency
Clock Divide
Mode
CDIV SCR SCO
Frequency
0 MCLK Divide by 1 1
1 MCLK
0 MCLK/2 Divide by 2 0
1 MCLK
An active low pulse of one SCO period on the data ready
output, DRDY, indicates a new conversion result is available at
the AD7763 serial data output,SDO.
Each bit of the new conversion result is clocked onto the SDO
line on the rising SCO edge and is valid on the falling SCO
edge. The SDO line is tri-stated when the serial data enable pin,
SDEN, is logic high.The 32-bit result consists of the 24 data bits
which, are followed by 8 status bits. These status bits are shown
in Table 7. The first three status bits. AD[2:0}, are the device
address bits. Table 15 contains descriptions of the other status
bits.
Table 7. Status Bits During Data Read
D7 D0
AD2 AD1 AD0 DValid Ovr LPwr FiltOk 0
The conversion result output on the SDO line is framed by the
frame synchronization output, FSO, which is sent logic low for
32 SCO cycles following the rising edge of the DRDY signal.
The AD7763 also features a serial data latch output, SDL, which
outputs a pulse every sixteen data bits. The SDL output offers an
alternative framing signal for serial transfers which require a
framing signal more frequent than every 32 bits.
Sharing The Serial Bus
The AD7763 functionality allows up to eight devices to share
the same serial bus, SDO. Devices sharing the serial output bus
are assigned addresses from 000 to 111.
For a part with an address of 0, the SDO line comes out of tri-
state on the first rising edge of SCO after the DRDY pulse and
returns to tri-state after the 32nd SCO falling edge plus a TBD
hold time. For the next device sharing the serial bus , address 1,
the SDO line will come out of tri-state on the 32nd SCO rising
edge for the next 32 SCO periods. This pattern will occur for
the rest of the proceeding devices sharing the serial bus.
Figure 4 shows an example lof four devices sharing the same
serial bus. Each of the devices is hardwired with a different
address form 0 to 3(ADR2:0). Device 0 outputs its serial word
on the SDO line during the first 32 SCO cycles. Device 1 then
outputs its serial word during the next 32 SCO cycles and so on
for the remaining devices 2 and 3.
To create the DRDY pulse as shown in Figure 4, (where four
devices are sharing the same serial line) the share bits of this
device, SH[2:0], should be set to 011, which, corresponds to 4
devices sharing the serial line. Each AD7763 device outputs its
own FSO signal.
The provision of two framing signals: DRDY and ,FSO ensures
that the AD7763 offers flexible data output framing options,
which, are further enhanced by the availability of the SDL
output. The user can select the framing output which bests suits
the application.
Writing To The AD7763
Figure 3 shows the AD7763 write operation. The serial writing
operation is synchronous to the SCO signal. The status of the
frame sync input, FSI , is checked on the falling edge of the SCO
signal. If the FSI line is low then the first data is latched in on
the next SCO falling edge.
The active edge of the FSI signal should be set to occur at a
position when the SCO signal is high or low and which also
allows set-up and hold time from the SCO falling edge to be
met. The width of the FSI signal may be set to between 1 and 32
SCO periods wide. A second or subsequent FSI falling edge
which occurs before 32 SCO periods have elapsed will be
ignored.
Figure 3 also shows the format for the serial data being written
to the AD7763. 32 bits are required for a write operation. The
first 16 bits are used to select the device and register address
which the data being read is intended. The second 16 bits
contain the data for the selected register. When using multiple
Preliminary Technical Data AD7763
Rev. PrD | Page 15 of 24
devices which share the same serial bus, all FSI and SDI pins
may be tied together and each device written to individually by
setting the appropriate address bits in the serial 32-bit word.
There is an exception to this in the case where all devices may
be written to at the same time by setting the ALL bit logic high.
Thus if this bit is set to logic high every device on the serial bus
will accept the data written regardless of the address bits. This
feature is particularly attractive, if, for example 4 devices were
being configured with the same user defined filter. Instead of
having to download the filter configuration four times, only one
write would be required.
Writing to AD7763 should be allowed at any time even while
reading a conversion result. It should be noted that after writing
to the devices, valid data will not be output until after the
settling time for the filter has elapsed. The DVALID status bit is
asserted at this point to indicate that the filter has settled and
that valid data is available at the output.
Reading Status and Other Registers
The AD7763 features a number of programmable registers. To
read back the contents of these registers or the status register,
the user must first write to the control register of the device
setting a bit corresponding to the register they wish to read. The
next read operation will then output the contents of the selected
register instead of a conversion result. More information on the
relevant bits in the control register is given in the AD7763
Registers section.
AD7763 Preliminary Technical Data
Rev. PrD | Page 16 of 24
CLOCKING THE AD7763
The AD7763 requires an external low jitter clock source. This
signal is applied to the MCLK and MCLKGND pins. An
internal clock signal (ICLK) is derived from the MCLK input
signal. This ICLK controls all the internal operation of the
AD7763. The maximum ICLK frequency is 20MHz but due to
an internal clock divider, a range of MCLK frequencies can be
used. There are three possibilities available to generate the
ICLK:
1. ICLK = MCLK (CDIV = 1)
2. ICLK = MCLK / 2 (CDIV = 0)
These options are pin selected (See Table 4 Pin Function
Descriptions for further details). If the user wishes to get output
data rates equal to those used in audio systems, a 12.288 MHz
ICLK frequency can be used. As shown in Table 5, output data
rates of 192, 96kHz and 48kHz are achievable with this ICLK
frequency. As mentioned previously, this ICLK frequency can
be derived from different MCLK frequencies.
The MCLK jitter requirements depend on a number of factors
and are given by the following equation:
20
)(
)(
102
dBSNR
IN
RMSj f
OSR
t
×××
=
π
OSR = Over-sampling ratio = ODR
fICLK
fIN = Maximum Input Frequency
SNR(dB) = Target SNR.
Taking an example from Table 5:
ODR = 625 kHz, fICLK = 20MHz, fIN (max) = 250 kHz, SNR =
108dB
pst RMSj 6.3
10102502
32
63
)( =
××××
=
π
This is the maximum allowable clock jitter for a full-scale
250kHz input tone with the given ICLK and Output Data Rate.
Taking a second example from Table 5:
ODR = 48kHz, fICLK = 12.288MHz, fIN (max) = 19.2kHz, SNR =
120dB
pst RMSj 133
10102.192
256
63
)( =
××××
=
π
The input amplitude also has an effect on these jitter figures. If,
for example, the input level was 3dB down from full-scale, the
allowable jitter would be increased by a factor of √2 increasing
the first example to 2.53ps RMS. This is due to the fact that the
maximum slew rate is reduced by a reduction in amplitude.
Figure 13 and Figure 14 illustrate this point showing the
maximum slew rate of a sine wave of the same frequency but
with different amplitudes.
Figure 13. Maximum Slew Rate of Sine Wave with Amplitude of 2V Pk-Pk
Figure 14. Maximum Slew Rate of Same Frequency Sine Wave with
Amplitude of 1V Pk-Pk
Preliminary Technical Data AD7763
Rev. PrD | Page 17 of 24
DRIVING THE AD7763
The AD7763 has an on-chip differential amplifier. This
amplifier will operate with a supply voltage (AVDD3) from 3V to
5.5V. For a 4.096V reference, the supply voltage must be 5V.
To achieve the specified performance in full power mode, the
differential amplifier should be configured as a first order anti-
alias filter as shown in Figure 15. Any additional filtering
should be carried out in previous stages using low noise, high-
performance op-amps such as the AD8021.
Suitable component values for the first order filter are listed in
Table 8. Using the first row as an example would yield a 10dB
attenuation at the first alias point of 19MHz.
A1
RIN
RIN
RFB
RFB
CFB
CFB
VIN-
VIN+
A
B
CS
Figure 15. Differential Amplifier Configuration
Table 8. Full Power Component Values
ODR VREF R
IN R
FB C
S C
FB
625kHz 4.096v
1k 655 5.6pF 33pF
625kHz 2.5v TBD TBD TBD TBD pF
48kHz 4.096v
TBD TBD TBD TBD pF
48kHz 2.5v TBD TBD TBD TBD pF
Figure 16 shows the signal conditioning that occurs using the
circuit in Figure 15 with a ±2.5V input signal biased around
ground using the component values and conditions in the first
row of Table 8. The differential amplifier will always bias the
output signal to sit on the optimum common mode of VREF/2, in
this case 2.048V. The signal is also scaled to give the maximum
allowable voltage swing with this reference value. This is
calculated as 80% of VREF, i.e. 0.8 × 4.096V 3.275V peak to
peak on each input.
0V
+2.5V
-2.5V
0V
+2.5V
-2.5V
+2.048V
0.410V
+3.685V
VIN+
VIN-
+2.048V
0.410V
+3.685V
A
B
Figure 16. Differential Amplifier Signal Conditioning
To obtain maximum performance from the AD7763, it is
advisable to drive the ADC with differential signals. However, it
is possible to drive the AD7763 with a single ended signal once
the common mode of the signal is within the range of +0.7V to
+2.1V with VDD3 = 5V or +0.7 to +1.25V with VDD3 = 3.3V. In
this case the on-chip differential amplifier can be used to
convert the signal from single-ended to differential before being
fed into the modulator inputs. Figure 17 shows how a bipolar
single-ended signal biased around ground can be used to drive
the AD7763 with the use of an external op-amp such as the
AD8021.
A1
RIN
RIN
RFB
RFB
CFB
CFB
VIN-
VIN+
AD8021
R
VIN
2R
2R
CS
Figure 17. Single Ended to Differential Conversion
If using a 4.096V reference, a supply must be provided to the
reference buffer (AVDD4). If using a 2.5V reference, a 3.3V
supply must be provided to AVDD4 .
AD7763 Preliminary Technical Data
Rev. PrD | Page 18 of 24
USING THE AD7763
The following is the recommended sequence for powering up
and using the AD7763.
1. Apply Power
2. Start clock oscillator, applying MCLK
3. Ta ke RESET low for a minimum of 1 MCLK cycle
4. Wait a minimum of 2 MCLK cycles after RESET has
been released.
5. Write to Control Register 2 to power up the ADC and
the differential amplifier as required.
6. Write to Control Register 1 to set up the Output Data
Rate.
7. In circumstances where multiple parts are being
synchronized, a SYNC pulse must be applied to the
parts, otherwise no SYNC pulse is required.
Conditions for applying the SYNC pulse:
(a) The issue of a SYNC pulse to the part must
not coincide with a write to the part.
(b) The SYNC pulse should be applied a
minimum of 2.5 ICLK cycles after the CS
signal for the previous write to the part has
returned logic high.
(c) Ensure that the SYNC pulse is taken low for
a minimum of 2.5 ICLK cycles.
Data can now be read from the part using the default filter,
offset, gain, and over range threshold values. The
conversion data read will not be valid however until the
settling time of the filter has passed. When this has
occurred, the DVALID bit read will be set indicating that
the data is indeed valid.
The user can now download their own filter if required
(see Downloading a User-Defined Filter). Values for gain,
offset and over range threshold registers can be written or
read at this stage.
BIAS RESISTOR SELECTION
The AD7763 requires a resistor to be connected between the
RBIAS pin and AGND. The value for this resistor is dependant on
the reference voltage being applied to the device. The resistor
value should be selected to give a current of 25µA through the
resistor to ground. For a 2.5V reference voltage, the correct
resistor value is 100k and for a 4.096V reference, 160k.
Preliminary Technical Data AD7763
Rev. PrD | Page 19 of 24
PROGRAMMABLE FIR FILTER
As previously mentioned, the third FIR filter on the AD7763 is
user programmable. The default coefficients that are loaded on
reset are given in Table 9. This gives a frequency response
shown in Figure 18. The frequencies quoted in Figure 18 scale
directly with the Output Data Rate.
Table 9. Default Filter Coefficients
# Dec. Value Hex
Value # Dec.
Value Hex Value
0 53656736 332BCA0 24 700847 AB1AF
1 25142688 17FA5A0 25 -70922 401150A
2 -4497814 444A196 26 -583959 408E917
3 -11935847 4B62067 27 -175934 402AF3E
4 -1313841 4140C31 28 388667 5EE3B
5 6976334 6A734E 29 294000 47C70
6 3268059 31DDDB 30 -183250 402CBD2
7 -3794610 439E6B2 31 -302597 4049E05
8 -3747402 4392E4A 32 16034 3EA2
9 1509849 1709D9 33 238315 3A2EB
10 3428088 344EF8 34 88266 158CA
11 80255 1397F 35 -143205 4022F65
12 -2672124 428C5FC 36 -128919 401F797
13 -1056628 4101F74 37 51794 CA52
14 1741563 1A92FB 38 121875 1DC13
15 1502200 16EBF8 39 16426 402A
16 -835960 40CC178 40 -90524 401619C
17 -1528400 4175250 41 -63899 400F99B
18 93626 16DBA 42 45234 B0B2
19 1269502 135EFE 43 114720 1C020
20 411245 6466D 44 102357 18FD5
21 -864038 40D2F26 45 52669 CDBD
22 -664622 40A242E 46 15559 3CC7
23 434489 6A139 47 1963 7AB
The default filter should be sufficient for almost all applications.
It is a standard brick wall filter with a symmetrical impulse
response. The default filter has a length of 96, is non-aliasing
with 120dB of attenuation at Nyquist. This filter not only
performs signal anti-aliasing but also suppresses out-of-band
quantization noise produced by the A-D conversion process.
Any significant relaxation in the stop-band attenuation or
transition band width relative to the default filter may result in a
failure to meet the SNR specifications.
If a user does wish to create their own filter then the following
should be noted:
The filter must be even, symmetrical FIR.
The coefficients are in sign-and-magnitude format
with 26 magnitude bits and sign coded as positive=0.
The filter length must be between 12 and 96 in steps of
12.
As the filter is symmetrical, the number of coefficients
that must be downloaded will be half the filter length.
The default filter coefficients are an example of this
with only 48 coefficients listed for a 96-tap filter.
Coefficients are written from the center of impulse
response (adjacent to the point of symmetry)
outwards.
The coefficients are scaled so that the in-band gain of
the filter is equal to 134217726 with the coefficients
rounded to the nearest integer. For a low pass filter
this is the equivalent of having the coefficients sum
arithmetically (including sign) to +67108863 (0x3FF
FFFF) positive value over the half-impulse-response
coefficient set (max 48 coefficients). Any deviation
from this will result in a gain error being introduced.
Figure 18. Default Filter Frequency Response (625kHz ODR)
The procedure for downloading a user-defined filter is detailed
in the Downloading a User-Defined Filter section.
AD7763 Preliminary Technical Data
Rev. PrD | Page 20 of 24
DOWNLOADING A USER-DEFINED FILTER
As previously mentioned, the filter coefficients are 27 bits in
length; one sign and 26 magnitude bits. Since the AD7763 has a
16-bit parallel bus, the coefficients are padded with 5 MSB zeros
to generate a 32-bit word and split into two 16-bit words for
downloading. The first 16-bit word for each coefficient becomes
(00000, Sign bit, Magnitude[25:16]), while the second word
becomes (Magnitude [15:0]). To ensure that a filter is
downloaded correctly, a checksum must also be generated and
downloaded following the final coefficient. The checksum is a
16-bit word generated by splitting each 32-bit word mentioned
above into 4 bytes and summing all bytes from all coefficients
up to a maximum of 192 bytes (48 coefficients × 4 bytes). The
same checksum is generated internally in the AD7763 and
compared with the checksum downloaded. The DL_OK bit in
the Status Register is set if these two checksums agree. The
following is the procedure for downloading a user filter:
1. Write to Control Register 1 setting the DL_Filt bit and
also the correct filter length bits corresponding to the
length of the filter about to be downloaded (See Table
10).
2. Write the first half of the current coefficient data
(00000, Sign bit, Magnitude[25:16]). The first
coefficient to be written must be the one adjacent to
the point of filter symmetry.
3. Write the second half of the current coefficient data
(Magnitude [15:0]).
4. Repeat Steps 2 and 3 for each coefficient.
5. Write the 16-bit checksum.
6. There are two methods to verify that the filter
coefficients have been downloaded correctly:
a. Read the Status Register checking the
DL_OK bit.
b. Start reading data and observe the status of
the DL_OK bit.
Table 10. Filter Length Values
FLEN[3:0] Num Coeffs Filter Length
0000 Default Default
0001 6 12
0011 12 24
0101 18 36
0111 24 48
1001 30 60
1011 36 72
1101 42 84
1111 48 96
It should be borne in mind that since the user coefficients are
stored in RAM, they will be cleared after a RESET operation or
a loss of power.
EXAMPLE FILTER DOWNLOAD
The following is an example of downloading a short user
defined filter with 24-taps. The frequency response is shown in
Figure 19.
-80
-70
-60
-50
-40
-30
-20
-10
0
10
0 100 200 300 400 500 600
Frequency (kHz)
Amplitude (dB)
Figure 19. 24-Tap FIR Frequency Response
The coefficients for the filter are listed in Table 11. The
coefficients are shown from the center of symmetry outwards.
The raw coefficients were generated using a commercially
available filter design tool and scaled appropriately so their sum
equals +67108863 (0x3FF FFFF).
Table 11. 24-Tap FIR Coefficients
Coeff Raw Scaled
1 0.365481974 53188232
2 0.201339905 29300796
3 0.009636604 1402406
4 -0.075708848 -11017834
5 -0.042856209 -6236822
6 0.019944246 2902466
7 0.036437914 5302774
8 0.007592007 1104856
9 -0.021556583 -3137108
10 -0.024888355 -3621978
11 -0.012379538 -1801582
12 -0.001905756 -277343
Preliminary Technical Data AD7763
Rev. PrD | Page 21 of 24
Table 12 shows the Hex values (in sign and magnitude format)
that are downloaded to the AD7763 to realize this filter. The
table is also split into the bytes which are all summed to
produce the checksum. The checksum generated from these
coefficients is 0x0E6B.
Table 12. Filter Hex Values
Coeff Word 1 Word 2
Byte 1 Byte 2 Byte 3 Byte 4
1 03 2B 96 88
2 01 BF 18 3C
3 00 15 66 26
4 04 A8 1E 6A
5 04 5F 2A 96
6 00 2C 49 C2
7 00 50 E9 F6
8 00 10 DB
D8
9 04 2F DE 54
10 04 37 44 5A
11 04 1B 7D 6E
12 04 04 3B 5F
What follows is a list of 16-bit words that the user would write
to the AD7763 to set up the ADC and download this filter
assuming an output data rate of 625kSPS has already been
selected.
0x0001 Address of Control Register 1
0x8079 Control Reg Data; DL Filter, Set Filter Length = 24, Set
Output Data Rate = 625kHz
0x032B First Coefficient, Word 1
0x9688 First Coefficient, Word 2
0x01BF Second Coefficient, Word 1
0x183C Second Coefficient, Word 2
… …
0x0404 Twelfth (Final) Coefficient, Word 1
0x3B5F Final Coefficient, Word 2
0x0E6B Checksum
Wait TBD xS for AD7763 to fill remaining unused coefficients
with zeros.
0x0001 Address of Control Register
0x0879 Control Reg Data; Set Read Status and maintain filter
length and decimation settings.
Read contents of Status Register. Check Bit 7 (DL_OK) to
determine that the filter was downloaded correctly.
AD7763 Preliminary Technical Data
Rev. PrD | Page 22 of 24
AD7763 REGISTERS
The AD7763 has a number of user-programmable registers. The control registers are used to set the decimation rate, the filter
configuration, the clock divider etc. There are also digital gain, offset and over-range threshold registers. Writing to these registers
involves writing the register address first, then a 16-bit data word. Register Addresses, details of individual bits and default values are
given here.
Table 13. Control Register 1 (Address 0x0001, Default Value 0x001A)
MSB LSB
DL
Filt
RD
Ovr
RD
Gain
RD
Off
RD
Stat 0 SYNC FLEN3 FLEN2 FLEN1 FLEN0 BYP F3 BYP F1 DEC2 DEC1 DEC0
Bit Mnemonic Comment
15 DL Filt8 Download Filter. Before downloading a user defined filter, this bit must be set. The Filter Length bits must also be set at
this time. The write operations that follow will be interpreted as the user coefficients for the FIR filter until all the
coefficients and the checksum have been written.
14 RD Ovr8,9 Read Overrange. If this bit has been set, the next read operation will output the contents of the Overrange Threshold
Register instead of a conversion result.
13 RD Gain8,9 Read Gain. If this bit has been set, the next read operation will output the contents of the digital Gain Register.
12 RD Off8,9 Read Offset. If this bit has been set, the next read operation will output the contents of the digital Offset Register.
11 RD Stat8,9 Read Status. If this bit has been set, the next read operation will output the contents of the Status Register.
10 0
9 SYNC8 Synchronize. Setting this bit will initiate in internal synchronisation routine. Setting this bit simultaneously on multiple
devices will synchronize all filters.
8-5 FLEN3:0 Filter Length Bits. These bits must be set when the DL Filt bit is set and before a user defined filter is downloaded.
4 BYP F3 Bypass Filter 3. If this bit is a 0, Filter 3 (Programmable FIR) will be bypassed.
3 BYP F1 Bypass Filter 1. If this bit is a 0, Filter 1 will be bypassed. This should only occur when the user requires unfiltered
modulator data to be output.
2-0 DEC2:0 Decimation Rate. These bits set the decimation rate of Filter 2. All zeros implies that the filter is bypassed. A value of 1
corresponds to 2x decimation, a value of 2 corresponds to 4x and so on up to the maximum value of 5, corresponding
to 32x decimation.
8 Bits 15-9 are all self clearing bits.
9 Only one of the bits may be set in any write operation as they all determine the contents of the next operation.
Table 14. Control Register 2 (Address 0x0002, Default Value 0x009B)
MSB LSB
0 0 0 0 0 0 0 0 0 0 0 0 PD LPWR 1 D1PD
Bit Mnemonic Comment
3 PD Power Down. Setting this bit powers down the AD7763 reducing the power consumption to TBD µW.
2 LPWR Low Power. If this bit is set, the AD7763 is operating in a low power mode. The power consumption is reduced for a 3dB
reduction in noise performance.
1 Write a ‘1’ to this bit.
0 D1PD Differential Amplifier Power Down. Setting this bit powers down the on-chip differential amplifier.
Preliminary Technical Data AD7763
Rev. PrD | Page 23 of 24
Table 15. Status Register (Read Only)
MSB LSB
PA RT
1
PA RT
0
DIE
2
DIE
1
DIE
0 DVALID LPWR OVR DL
OK
Filter
OK
U
Filter BYP F3 BYP F1 DEC2 DEC1 DEC0
Bit Mnemonic Comment
15,14 PART1:0 Part Number. These bits will be constant for the AD7763.
13-11 DIE2:0 Die Number. These bits will reflect the current AD7763 die number for identification purposes within a system.
10 DVALID Data Valid. This bit corresponds to the DVALID bit in the status word output in the second 16-bit read operation.
9 LPWR Low Power. If the AD7763 is operating in Low Power Mode, this bit is set to a 1.
8 OVR If the current analog input exceeds the current overrange threshold, this bit will be set.
7 DL OK When downloading a user filter to the AD7763, a checksum is generated. This checksum is compared to the one
downloaded following the coefficients. If these checksums agree, this bit is set.
6 Filter OK
When a user-defined filter is in use, a checksum is generated when the filter coefficients pass through the filter. This
generated checksum is compared to the one downloaded. If they match, this bit is set.
5 U Filter If a user-defined filter is in use, this bit is set.
4 BYP F3 Bypass Filter 3. If Filter 3 is bypassed by setting the relevant bit in Control Register 1, this bit is also set.
3 BYP F1 Bypass Filter 1. If Filter 1 is bypassed by setting the relevant bit in Control Register 1, this bit is also set.
2-0 DEC2:0 Decimation Rate. These correspond to the bits set in Control Register 1.
NON BIT-MAPPED REGISTERS
Offset Register (Address 0x0003, Default Value 0x0000)
The Offset Register uses 2’s Complement notation and is scaled such that 0x7FFF (maximum positive value) and 0x8000 (maximum
negative value) correspond to an offset of +0.78125% and -0.78125% respectively. Offset correction is applied after any gain correction.
Using the default gain value of 1.25 and assuming a reference voltage of 4.096V, the offset correction range is approximately ±25mV.
Gain Register (Address 0x0004, Default Value 0xA000)
The Gain Register is scaled such that 0x8000 corresponds to a gain of 1.0. The default value of this register is 1.25 (0xA000). This gives a
full scale digital output when the input is at 80% of VREF. This ties in with the maximum analog input range of ±80% of VREF Pk-Pk.
Over Range Register (Address 0x0005, Default Value 0xCCCC)
The Over Range register value is compared with the output of the first decimation filter to obtain an overload indication with minimum
propagation delay. This is prior to any gain scaling or offset adjustment. The default value is 0xCCCC which corresponds to 80% of VREF
(the maximum permitted analog input voltage) Assuming VREF = 4.096V, the bit will then be set when the input voltage exceeds
approximately 6.55v pk-pk differential. Note that the over-range bit is also set immediately if the analog input voltage exceeds 100% of
VREF for more than 4 consecutive samples at the modulator rate.
AD7763 Preliminary Technical Data
Rev. PrD | Page 24 of 24
OUTLINE DIMENSIONS
1
12
13
24
25
36
37
48
BOTTOM
VIEW
7.0 (0.276) BSC SQ
5.25 (0.207)
5.10 (0.201) SQ
4.95 (0.195)
0.60 (0.024)
0.42 (0.017)
0.24 (0.009)
0.50 (0.020)
0.40 (0.016)
0.30 (0.012)
PIN 1
INDICATOR
TOP
VIEW
6.75 (0.266)
BSC SQ
5.5 (0.217)
REF
0.25 (0.010)
MIN
0.60 (0.024)
0.42 (0.017)
0.24 (0.009)
0.70 (0.028) MAX
0.65 (0.026) NOM0.05 (0.002)
0.01 (0.0004)
0.0 (0.0)
0.20 (0.008)
REF
0.50 (0.020)
BSC
12°MAX
0.90 (0.035) MAX
0.85 (0.033) NOM
0.30 (0.012)
0.23 (0.009)
0.18 (0.007)
Figure 20. 48-Lead Frame Chip Scale Package [LFCSP] (CP-48)—Dimensions shown in millimeters
1
16
17
33
32
48
4964
12.0(0.47) BSC
10.0(0.39) BSC
0.50 (0.02)
BSC
0.22 ± 0.05
(0.0087 ± 0.002)
SEATING
PLANE
1.60 (0.063) MAX
0.60 ± 0.15
(0.024 ± 0.006) 12o
TYP
0o
3.5o ± 3.5o
0.15(0.006)
0.05(0.002)
TOP VIEW 6.0(0.235)
BSC
Figure 21. 64-Lead Thin Quad Flat Pack (Exposed Paddle) [TQFP] (SV-64)—Dimensions shown in millimeters
ORDERING GUIDE
Model Temperature Range Package Description Package Option
AD7763BCP –40°C to +85°C Lead Frame Chip Scale Package CP-48
AD7763BSV –40°C to +85°C Thin Quad Flat Pack, Exposed Paddle SV-64
© 2005 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
Printed in the U.S.A.
PR05476-0-4/05(PrD)