2.5 MSPS, 24-Bit, 100 dB
Sigma-Delta ADC with On-Chip Buffer
AD7760
Rev. A
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: 781.461.3113 ©2006 Analog Devices, Inc. All rights reserved.
FEATURES
120 dB dynamic range at 78 kHz output data rate
100 dB dynamic range at 2.5 MHz output data rate
112 dB SNR at 78 kHz output data rate
100 dB SNR at 2.5 MHz output data rate
2.5 MHz maximum fully filtered output word rate
Programmable oversampling rate (8× to 256×)
Fully differential modulator input
On-chip differential amplifier for signal buffering
Low-pass finite impulse response (FIR) filter with default or
user-programmable coefficients
Modulator output mode
Overrange alert bit
Digital offset and gain correction registers
Filter bypass modes
Low power and power-down modes
Synchronization of multiple devices via SYNC pin
APPLICATIONS
Data acquisition systems
Vibration analysis
Instrumentation
FUNCTIONAL BLOCK DIAGRAM
AD7760
V
IN
+
V
IN
–
AV
DD
1
AGND
MCLK
DGND
V
DRIVE
AV
DD
2
AV
DD
3
AV
DD
4
DV
DD
DECAPA/B
R
BIAS
DB0 TO DB15CS DRDYRD/WR
CONTROL LOGIC
I/O
OFFSET AND GAIN
REGISTERS
DIFF MULTIBIT
Σ-Δ
MODULATOR
RECONSTRUCTION
V
REF+
FIR FILTER
ENGINE
PROGRAMMABLE
DECIMATION
BUF
SYNC
RESET
0
4975-001
Figure 1.
GENERAL DESCRIPTION
The AD7760 is a high performance, 24-bit Σ- analog-to-digital
converter (ADC). It combines wide input bandwidth and high
speed with the benefits of Σ- conversion to achieve a perfor-
mance of 100 dB SNR at 2.5 MSPS, making it ideal for high
speed data acquisition. Wide dynamic range combined with
significantly reduced antialiasing requirements simplify the
design process. An integrated buffer to drive the reference, a
differential amplifier for signal buffering and level shifting, an
overrange flag, internal gain and offset registers, and a low-pass
digital FIR filter make the AD7760 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 for the application. The
AD7760 is ideal for applications demanding high SNR
without a complex front-end signal processing design.
The differential input is sampled at up to 40 MSPS by an analog
modulator. The modulator output is processed by a series of low-
pass filters, with the final filter 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 AD7760.
The reference voltage supplied to the AD7760 determines the
analog input range. With a 4 V reference, the analog input range
is ±3.2 V differential biased around a common mode of 2 V.
This common-mode biasing can be achieved using the on-chip
differential amplifier, further reducing the external signal
conditioning requirements.
The AD7760 is available in an exposed paddle, 64-lead TQFP
and is specified over the industrial temperature range from
−40°C to +85°C.
Table 1. Related Devices
Part No. Description
AD7762 24-bit, 625 kSPS, 109 dB, Σ-∆ parallel interface
AD7763 24-bit, 625 kSPS, 109 dB, Σ-∆ serial interface
AD7760
Rev. A | Page 2 of 36
TABLE OF CONTENTS
Features .............................................................................................. 1
Applications....................................................................................... 1
Functional Block Diagram .............................................................. 1
General Description......................................................................... 1
Revision History ............................................................................... 3
Specifications..................................................................................... 4
Timing Specifications .................................................................. 6
Timing Diagrams.......................................................................... 7
Absolute Maximum Ratings............................................................ 8
ESD Caution.................................................................................. 8
Pin Configuration and Function Descriptions............................. 9
Terminology .................................................................................... 11
Typical Performance Characteristics ........................................... 12
Theory of Operation ...................................................................... 18
Modulator Data Output Mode...................................................... 19
Modulator Inputs........................................................................ 19
Modulator Data Output Scaling............................................... 19
Modulator Data Output Mode Interface..................................... 20
Clock Divide-by-1 Mode (CDIV = 1) ..................................... 20
Clock Divide-by-2 Mode (CDIV = 0) ..................................... 20
Using the AD7760 in Modulator Output Mode..................... 21
AD7760 Interface............................................................................ 22
Reading Data...............................................................................22
Reading Status and Other Registers......................................... 22
Sharing the Parallel Bus............................................................. 22
Synchronization.......................................................................... 22
Writing to the AD7760.............................................................. 23
Clocking the AD7760 .................................................................... 24
Buffering the MCLK signal....................................................... 24
MCLK Jitter Requirements ....................................................... 24
Driving the AD7760....................................................................... 26
Using the AD7760...................................................................... 27
Decoupling and Layout Recommendations................................ 28
Supply Decoupling..................................................................... 29
Additional Decoupling.............................................................. 29
Reference Voltage Filtering ....................................................... 29
Differential Amplifier Components ........................................ 29
Bias Resistor Selection ............................................................... 29
Layout Considerations............................................................... 29
Exposed Paddle........................................................................... 29
Programmable FIR Filter............................................................... 30
Downloading a User-Defined Filter ............................................ 31
Example Filter Download ......................................................... 31
AD7760 Registers........................................................................... 33
Control Register 1—Address 0x0001 ...................................... 33
Control Register 2—Address 0x0002 ...................................... 33
Status Register (Read Only) ...................................................... 34
Offset Register—Address 0x0003............................................. 34
Gain Register—Address 0x0004............................................... 34
Overrange Register—Address 0x0005..................................... 34
Outline Dimensions....................................................................... 35
Ordering Guide .......................................................................... 35
AD7760
Rev. A | Page 3 of 36
REVISION HISTORY
8/06—Rev. 0 to Rev. A
Updated Package Option...................................................Universal
Change to Features............................................................................1
Changes to Specifications.................................................................4
Changes to Absolute Maximum Ratings........................................8
Changes to Terminology Section ..................................................11
Added Figure 36 Through Figure 39............................................17
Added Modulator Data Output Mode Section ...........................19
Added Figure 41 Through Figure 47............................................19
Added Modulator Data Output Mode Interface Section...........20
Changes to Reading Data Section.................................................22
Added Synchronization Section....................................................22
Changes to Clocking the AD7760 Section...................................24
Added Buffering the MCLK Signal Section.................................24
Added MCLK Jitter Requirements Heading ...............................24
Changes to Driving the AD7760 Section.....................................26
Changes to Figure 51 ......................................................................26
Added Figure 52 ..............................................................................26
Changes to Figure 55 ......................................................................28
Changes to Figure 56 ......................................................................29
Added Exposed Paddle Section.....................................................29
Change to Control Register 2—Address 0x0002 Section ..........33
Changes to Status Register (Read Only) Section ........................34
7/05—Revision 0: Initial Version
AD7760
Rev. A | Page 4 of 36
SPECIFICATIONS
AVDD1 = DVDD = VDRIVE = 2.5 V, AVDD2 = AVDD3 = AVDD4 = 5 V, VREF+ = 4.096 V, MCLK amplitude = 5 V, TA = 25°C, normal mode, using
the on-chip amplifier with components as shown in Table 8, unless otherwise noted.1
Table 2.
Parameter Test Conditions/Comments Specification Unit
DYNAMIC PERFORMANCE
Decimate by 256 MCLK = 40 MHz, ODR = 78 kHz, fIN = 1 kHz
Dynamic Range Modulator inputs shorted 119 dB min
120.5 dB typ
Signal-to-Noise Ratio (SNR)2Input amplitude = −0.5 dBFS 112 dB typ
Input amplitude = −60 dBFS 59 dB typ
Spurious-Free Dynamic Range (SFDR) Nonharmonic, input amplitude = −6 dBFS 126 dBc typ
Input amplitude = −60 dBFS 77 dBc typ
Total Harmonic Distortion (THD) Input amplitude = −0.5 dBFS −105 dB typ
Input amplitude = −6 dBFS −106 dB typ
Input amplitude = −60 dBFS −75 dB typ
Decimate by 32 MCLK = 40 MHz, ODR = 625 kHz, fIN =100 kHz
Dynamic Range Modulator inputs shorted 108 dB min
109.5 dB typ
Signal-to-Noise Ratio (SNR)2 Input amplitude = −0.5 dBFS 107 dB typ
Spurious-Free Dynamic Range (SFDR) Nonharmonic, input amplitude = −6 dBFS 120 dBc typ
Total Harmonic Distortion (THD) Input amplitude = −0.5 dBFS −105 dB typ
Input amplitude = −6 dBFS −106 dB typ
Decimate by 8 MCLK = 40 MHz, ODR = 2.5 MHz
Dynamic Range Modulator inputs shorted 99 dB min
100.5 dB typ
Signal-to-Noise Ratio (SNR)2 f
IN = 1 kHz, input amplitude = −0.5 dBFS 100 dB typ
f
IN = 100 kHz, input amplitude = −0.5 dBFS 99 dB typ
f
IN = 1 MHz, input amplitude = −0.5 dBFS 98 dB typ
Spurious-Free Dynamic Range (SFDR) Nonharmonic, fIN = 100 kHz, input amplitude = −6 dBFS 120 dBc typ
Nonharmonic, fIN = 1 MHz, input amplitude = −6 dBFS 114 dBc typ
Total Harmonic Distortion (THD) Input amplitude = −0.5 dBFS, fIN = 100 kHz −103 dB typ
Input amplitude = −6 dBFS, fIN = 100 kHz −102 dB typ
IMD Second Order fIN A = 989.95 kHz, fIN B = 999.95 kHz −115 dB typ
IMD Third Order fIN A = 989.95 kHz, fIN B = 999.95 kHz −89 dB typ
DC ACCURACY
Resolution 24 Bits
Differential Nonlinearity Guaranteed monotonic to 24 bits
Integral Nonlinearity 0.00076 % typ
Zero Error 0.014 % typ
0.02 % max
Gain Error 0.016 % typ
Zero Error Drift 0.00001 % FS/°C typ
Gain Error Drift 0.0002 % FS/°C typ
DIGITAL FILTER RESPONSE
Decimate by 8
Group Delay MCLK = 40 MHz 12 µs typ
Decimate by 32
Group Delay MCLK = 40 MHz 47 µs typ
Decimate by 256
Group Delay MCLK = 40 MHz 358 µs typ
AD7760
Rev. A | Page 5 of 36
Parameter Test Conditions/Comments Specification Unit
ANALOG INPUT
Differential Input Voltage VIN(+) – VIN(−), VREF = 2.5 V ±2 V p-p
V
IN(+) – VIN(−), VREF = 4.096 V ±3.25 V p-p
Input Capacitance At internal buffer inputs 5 pF typ
At modulator inputs 55 pF typ
REFERENCE INPUT/OUTPUT
VREF Input Voltage VDD3 = 3.3 V ± 5% +2.5 V max
V
DD3 = 5 V ± 5% +4.096 V max
VREF Input DC Leakage Current ±6 µA max
VREF Input Capacitance 5 pF max
POWER DISSIPATION
Total Power Dissipation Normal mode 958 mW max
Low power mode 661 mW max
Standby Mode Clock stopped 6.35 mW max
POWER REQUIREMENTS
AVDD1 (Modulator Supply) ±5% +2.5 V
AVDD2 (General Supply) ±5% +5 V
AVDD3 (Differential Amplifier Supply) +3.15/+5.25 V min/max
AVDD4 (Reference Buffer Supply) +3.15/+5.25 V min/max
DVDD ±5% +2.5 V
VDRIVE +1.65/+2.7 V min/max
Normal Mode
AIDD1 (Modulator) 49/51 mA typ/max
AIDD2 (General)3 40/42 mA typ/max
AIDD4 (Reference Buffer) AVDD4 = 5 V 34/36 mA typ/max
Low Power Mode
AIDD1 (Modulator) 26/28 mA typ/max
AIDD2 (General)3 20/23 mA typ/max
AIDD4 (Reference Buffer) AVDD4 = 5 V 9/10 mA typ/max
AIDD3 (Differential Amplifier) AVDD3 = 5 V, both modes 41/44 mA typ/max
DIDD Both modes 63/70 mA typ/max
DIGITAL I/O
MCLK Input Amplitude4 5 V typ
Input Capacitance 7.3 pF typ
Input Leakage Current ±5 A max
Three-State Leakage Current (D15:D0) ±5 A max
VINH 0.7 × VDRIVE V min
VINL 0.3 × VDRIVE V max
VOH5 1.5 V min
VOH6 2.4 V typ
VOL4 0.1 V max
1 See the Terminology section.
2 SNR specifications in decibels are referred to a full-scale input, FS. Tested with an input signal at 0.5 dB below full scale, unless otherwise specified.
3 Current scales with ICLK frequency. See the Typical Performance Characteristics section.
4 Although the AD7760 can function with an MCLK amplitude of less than 5 V, this is the recommended amplitude to achieve the performance as stated.
5 Tested using the minimum VDRIVE voltage of 1.65 V with a 400 µA load current.
6 Tested using VDRIVE = 2.5 V with a 400 A load current.
AD7760
Rev. A | Page 6 of 36
TIMING SPECIFICATIONS
AVDD1 = DVDD = VDRIVE = 2.5 V, AVDD2 = AVDD3 = AVDD4 = 5 V, TA = 25°C, normal mode, unless otherwise noted.
Table 3.
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
t11, 20.5 × tICLK typ
DRDY pulse width
t2 10 ns min
DRDY falling edge to CS falling edge
t3 3 ns min
RD/WR setup time to CS falling edge
t4 (0.5 × tICLK) + 16 ns max Data access time
t5 t
ICLK min
CS low read pulse width
t6 t
ICLK min
CS high pulse width between reads
t7 3 ns min
RD/WR hold time to CS rising edge
t8 11 ns max Bus relinquish time
t92 0.5 × tICLK typ
DRDY high period
t102 0.5 × tICLK typ
DRDY low period
t11 (0.5 × tICLK) + 16 ns max Data access time
t123, 423 ns min
Data valid prior to DRDY rising edge
t133, 419 ns min
Data valid after DRDY rising edge
t14 11 ns max Bus relinquish time
t15 4 × tICLK min
CS low write pulse width
t16 4 × tICLK min
CS high period between address and data
t17 5 ns min Data setup time
t18 0 ns min Data hold time
t194, 523 ns min
Data valid prior to MCLK falling edge while DRDY is logic low
t204, 519 ns min
Data valid after MCLK falling edge while DRDY is logic low
1 tICLK = 1/fICLK.
2 When ICLK = MCLK, DRDY pulse width depends on the mark-space ratio of applied MCLK.
3 Valid when using the modulator output mode with CDIV = 1.
4 See the Modulator Data Output Mode section for timing diagrams.
5 Valid when using the modulator output mode with CDIV = 0.
AD7760
Rev. A | Page 7 of 36
TIMING DIAGRAMS
DATA MSW LSW + STATUS
t
5
t
8
t
7
t
6
t
3
t
4
t
2
t
1
D[0:15]
CS
RD/WR
DRDY
04975-002
Figure 2. Filtered Output—Parallel Interface Timing Diagram
t
15
D[0:15]
CS
RD/WR
t
16
t
17
t
18
REGISTER ADDRESS REGISTER DATA
04975-004
Figure 3. AD7760 Register Write
AD7760
Rev. A | Page 8 of 36
ABSOLUTE MAXIMUM RATINGS
TA = 25°C, unless otherwise noted.
Table 4.
Parameters Rating
AVDD1 to GND −0.3 V to +3 V
AVDD2:AVDD4 to GND −0.3 V to +6 V
DVDD to GND −0.3 V to +3 V
VDRIVE to GND −0.3 V to +3 V
VIN+, VIN– to GND1 −0.3 V to +6 V
VINA+, VINA− to GND1 −0.3 V to +6 V
Digital Input Voltage to GND2 −0.3 V to DVDD + 0.3 V
MCLK to MCLKGND −0.3 V to +6 V
VREF+ to GND3 −0.3 V to AVDD4 + 0.3 V
AGND to DGND −0.3 V to +0.3 V
Input Current to Any Pin Except
Supplies4
±10 mA
Operating Temperature Range
Commercial −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
Lead Temperature, Soldering
Vapor Phase (60 sec) 215°C
Infrared (15 sec) 220°C
ESD 600 V
1 Absolute maximum voltage for VIN−, VIN+ and VINA−, VINA+ is 6.0 V or AVDD3 + 0.3 V,
whichever is lower.
2 Absolute maximum voltage on digital inputs is 3.0 V or DVDD + 0.3 V,
whichever is lower.
3 Absolute maximum voltage on VREF+ input is 6.0 V or AVDD4 + 0.3 V,
whichever is lower.
4 Transient currents of up to 200 mA do not cause SCR latch-up.
Stresses above those listed under Absolute Maximum Ratings
may cause permanent damage to the device. This is a stress
rating only; functional operation of the device at these or any
other conditions above those listed in the operational sections
of this specification is not implied. Exposure to absolute
maximum rating conditions for extended periods may affect
device reliability.
ESD CAUTION
AD7760
Rev. A | Page 9 of 36
PIN CONFIGURATION AND FUNCTION DESCRIPTIONS
64
DGND
63
V
DRIVE
62
DGND
61
DB0
60
DB1
59
DB2
58
DB3
57
DB4
56
DB5
55
DB6
54
DB7
53
DGND
52
DB8
51
DB9
50
DB10
49
DB11
47
DB13
46
DB14
45
DB15
42
DGND
43
DGND
44
V
DRIVE
48
DB12
41
DV
DD
40
CS
39
RD/WR
37
RESET
36
SYNC
35
DGND
34
AGND1
33
AV
DD
1
38
DRDY
2
MCLKGND
3
MCLK
4
AV
DD
2
7
AGND1
6
AV
DD
1
5
AGND2
1
DGND
8
DECAPA
9
REFGND
10
V
REF+
12
AV
DD
4
13
AGND2
14
AV
DD
2
15
AV
DD
2
16
AGND2
11
AGND4
PIN 1
17
R
BIAS
18
AGND2
19
V
IN
A+
20
V
IN
A–
21
V
OUT
A–
22
V
OUT
A+
23
AGND3
24
AV
DD
3
25
V
IN
+
26
V
IN
–
27
AV
DD
2
28
AGND2
29
AGND3
30
DECAPB
31
AGND3
32
AGND3
AD7760
TOP VIEW
(Not to Scale)
0
4975-005
Figure 4. 64-Lead TQFP Pin Configuration
Table 5. Pin Function Descriptions
Pin No. Mnemonic Description
6, 33 AVDD1 2.5 V Power Supply for Modulator. These pins should be decoupled to AGND1 (Pin 7 and Pin 34, respectively)
with 100 nF and 10 µF capacitors on each pin. See the Decoupling and Layout Recommendations section
for details.
4, 14, 15, 27 AVDD2 5 V Power Supply. These pins should be decoupled to AGND2 (Pin 5 and Pin 13, with 100 nF capacitors
on each of Pin 4, Pin 14, and Pin 15). Pin 27 should be connected to Pin 14 via a 15 nH inductor. See the
Decoupling and Layout Recommendations section for details.
24 AVDD3 3.3 V to 5 V Power Supply for Differential Amplifier. This pin should be decoupled to AGND3 (Pin 23) with a
100 nF capacitor. See the Decoupling and Layout Recommendations section for details.
12 AVDD4 3.3 V to 5 V Power Supply for Reference Buffer. This pin should be decoupled to Pin 9 with a 10 nF capacitor
in series with a 10 Ω resistor.
7, 34 AGND1 Power Supply Ground for Analog Circuitry Powered by AVDD1.
5, 13, 16, 18, 28 AGND2 Power Supply Ground for Analog Circuitry Powered by AVDD2.
23, 29, 31, 32 AGND3 Power Supply Ground for Analog Circuitry Powered by AVDD3.
11 AGND4 Power Supply Ground for Analog Circuitry Powered by AVDD4.
9 REFGND Reference Ground. Ground connection for the reference voltage.
41 DVDD 2.5 V Power Supply for Digital Circuitry and FIR Filter. This pin should be decoupled to DGND with a 100 nF
capacitor.
44, 63 VDRIVE Logic Power Supply Input, 1.8 V to 2.5 V. The voltage supplied at these pins determines the operating
voltage of the logic interface. Both of these pins must be connected together and tied to the same supply.
Each pin should also be decoupled to DGND with a 100 nF capacitor.
1, 35, 42, 43,
53, 62, 64
DGND Ground Reference for Digital Circuitry.
19 VINA+ Positive Input to Differential Amplifier.
20 VINA− Negative Input to Differential Amplifier.
21 VOUTA− Negative Output from Differential Amplifier.
22 VOUTA+ Positive Output from Differential Amplifier.
25 VIN+ Positive Input to the Modulator.
26 VIN− Negative Input to the Modulator.
10 VREF+ Reference Input. The input range of this pin is determined by the reference buffer supply voltage (AVDD4).
See the Reference Voltage Filtering section for more details.
8 DECAPA Decoupling Pin. A 100 nF capacitor must be inserted between this pin and AGND.
AD7760
Rev. A | Page 10 of 36
Pin No. Mnemonic Description
30 DECAPB Decoupling Pin. A 33 pF capacitor must be inserted between this pin and AGND3.
17 RBIAS Bias Current Setting Pin. A resistor must be inserted between this pin and AGND. For more details, see the
Bias Resistor Selection section.
45 to 52,
54 to 61
DB15:DB8,
DB7:DB0
16-Bit Bidirectional Data Bus. These are three-state pins that are controlled by the CS pin and the RD/WR
pin. The operating voltage for these pins is determined by the VDRIVE voltage. See the Modulator Data
Output Mode and AD7760 Interface sections for more details.
37 RESET A falling edge on this pin resets all internal digital circuitry and powers down the part. Holding this pin low
keeps the AD7760 in a reset state.
3 MCLK
Master Clock Input. A low jitter, buffered digital clock must be applied to this pin. The output data rate
depends on the frequency of this clock. See the Clocking the AD7760 section for more details.
2 MCLKGND Master Clock Ground Sensing Pin.
36 SYNC Synchronization Input. A falling edge on this pin resets the internal filter. This can be used to synchronize
multiple devices in a system. See the Synchronization section for more details.
39 RD/WR Read/Write Input. This pin, in conjunction with the chip select pin, is used to read and write data to and
from the AD7760. If this pin is low when CS is low, a read takes place. If this pin is high when CS is low, a
write occurs. See the Modulator Data Output Mode and AD7760 Interface sections for more details.
38 DRDY Data Ready Output. Each time new conversion data is available, an active low pulse, ½ ICLK period wide, is
produced on this pin. See the Modulator Data Output Mode and AD7760 Interface sections for more details.
40 CS Chip Select Input. Used in conjunction with the RD/WR pin to read and write data from and to the AD7760.
See the Modulator Data Output Mode and AD7760 Interface sections for more details.
AD7760
Rev. A | Page 11 of 36
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. The value for SNR is
expressed in decibels.
Total Harmonic Distortion (THD)
THD is the ratio of the rms sum of harmonics to the fundamental.
For the AD7760, it is defined as
()
1
6
54
32
V
VVVVV
THD
22222
log20dB ++++
=
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.
Nonharmonic Spurious-Free Dynamic Range (SFDR)
SFDR is the ratio of the rms signal amplitude to the rms value
of the peak spurious spectral component, excluding harmonics.
Dynamic Range
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 the dynamic range is expressed in decibels.
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 0. For example, the second-
order terms include (fa + fb) and (fa − fb), and the third-order
terms include (2fa + fb), (2fa − fb), (fa + 2fb), and (fa − 2fb).
The AD7760 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, and 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 decibels.
Integral Nonlinearity (INL)
INL is the maximum deviation from a straight line passing
through the endpoints of the ADC transfer function.
Differential Nonlinearity (DNL)
DNL is the difference between the measured and the ideal
1 LSB change between any two adjacent codes in the ADC.
Zero Error
Zero error is the difference between the ideal midscale input
voltage (when both inputs are shorted together) and the actual
voltage producing the midscale output code.
Zero Error Drift
Zero error drift is the change in the actual zero error value due
to a temperature change of 1°C. It is expressed as a percentage
of full scale at room temperature.
Gain Error
The first transition (from 100 … 000 to 100 … 001) should
occur for an analog voltage ½ LSB above the nominal negative
full scale. The last transition (from 011 … 110 to 011 … 111)
should occur for an analog voltage 1½ 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.
Gain Error Drift
Gain error drift is the change in the actual gain error value due
to a temperature change of 1°C. It is expressed as a percentage
of full scale at room temperature.
AD7760
Rev. A | Page 12 of 36
TYPICAL PERFORMANCE CHARACTERISTICS
AVDD1 = DVDD = VDRIVE = 2.5 V, AVDD2 = AVDD3 = AVDD4 = 5 V, VREF = 4.096 V, TA = 25°C, normal mode, unless otherwise noted. All
FFTs are generated from 65,536 samples using a 7-term Blackman-Harris window.
04975-006
0 4000 8000 12000 16000 20000
–200
0
–25
–50
–75
–100
–125
–150
–175
24000
FREQUENCY (Hz)
AMPLITUDE (dB)
Figure 5. Normal Mode FFT, 1 kHz, −0.5 dB Input Tone, 256× Decimation
04975-007
0 4000 8000 12000 16000 20000
–200
0
–25
–50
–75
–100
–125
–150
–175
24000
FREQUENCY (Hz)
AMPLITUDE (dB)
Figure 6. Normal Mode FFT, 1 kHz, −6 dB Input Tone, 256× Decimation
04975-008
0 4000 8000 12000 16000 20000
–200
0
–25
–50
–75
–100
–125
–150
–175
24000
FREQUENCY (Hz)
AMPLITUDE (dB)
Figure 7. Normal Mode FFT, 1 kHz, −60 dB Input Tone, 256× Decimation
04975-009
0 4000 8000 12000 16000 20000
–200
0
–25
–50
–75
–100
–125
–150
–175
24000
FREQUENCY (Hz)
AMPLITUDE (dB)
Figure 8. Low Power FFT, 1 kHz, −0.5 dB Input Tone, 256× Decimation
04975-010
0 4000 8000 12000 16000 20000
–200
0
–25
–50
–75
–100
–125
–150
–175
24000
FREQUENCY (Hz)
AMPLITUDE (dB)
Figure 9. Low Power FFT, 1 kHz, −6 dB Input Tone, 256× Decimation
04975-011
0 4000 8000 12000 16000 20000
–200
0
–25
–50
–75
–100
–125
–150
–175
24000
FREQUENCY (Hz)
AMPLITUDE (dB)
Figure 10. Low Power FFT, 1 kHz, −60 dB Input Tone, 256× Decimation
AD7760
Rev. A | Page 13 of 36
04975-012
0 250 500 750 1000 1250
–200
0
–25
–50
–75
–100
–125
–150
–175
FREQUENCY (kHz)
AMPLITUDE (dB)
Figure 11. Normal Mode FFT, 100 kHz, −0.5 dB Input Tone, 8× Decimation
04975-013
0 250 500 750 1000 1250
–200
0
–25
–50
–75
–100
–125
–150
–175
FREQUENCY (kHz)
AMPLITUDE (dB)
Figure 12. Normal Mode FFT, 100 kHz, −6 dB Input Tone, 8× Decimation
04975-014
0 250 500 750 1000 1250
–200
0
–25
–50
–75
–100
–125
–150
–175
FREQUENCY (kHz)
AMPLITUDE (dB)
Figure 13. Normal Mode FFT, 1 MHz, −0.5 dB Input Tone, 8× Decimation
04975-015
0 250 500 750 1000 1250
–200
0
–25
–50
–75
–100
–125
–150
–175
FREQUENCY (kHz)
AMPLITUDE (dB)
Figure 14. Low Power FFT, 100 kHz, −0.5 dB Input Tone, 8× Decimation
04975-016
0 250 500 750 1000 1250
–200
0
–25
–50
–75
–100
–125
–150
–175
FREQUENCY (kHz)
AMPLITUDE (dB)
Figure 15. Low Power FFT, 100 kHz, −6 dB Input Tone, 8× Decimation
04975-017
0 250 500 750 1000 1250
–200
0
–25
–50
–75
–100
–125
–150
–175
FREQUENCY (kHz)
AMPLITUDE (dB)
Figure 16. Low Power FFT, 1 MHz, −0.5 dB Input Tone, 8× Decimation
AD7760
Rev. A | Page 14 of 36
0
–200
0 1250
04975-018
FREQUENCY (kHz)
AMPLITUDE (dB)
–25
–50
–75
–100
–125
–150
–175
250 500 750 1000
Figure 17. Normal Mode FFT, 1 MHz, −6 dB Input Tone, 8× Decimation
0
–200
0 1250
04975-019
FREQUENCY (kHz)
AMPLITUDE (dB)
–25
–50
–75
–100
–125
–150
–175
250 500 750 1000
TONE A: 999.75kHz
TONE B: 1.00025MHz
Figure 18. Normal Mode IMD, 1 MHz Center Frequency, 8× Decimation
0
–160
0 10000
04975-020
FREQUENCY (kHz)
AMPLITUDE (dB)
–20
–40
–60
–80
–100
–120
–140
2000 4000 6000 8000
TONE A: 999.75kHz
TONE B: 1.00025MHz
SECOND-ORDER IMD: –105.6dB
Figure 19. Normal Mode IMD, 1 MHz Center Frequency, 8× Decimation
0
–200
0 1250
04975-021
FREQUENCY (kHz)
AMPLITUDE (dB)
–25
–50
–75
–100
–125
–150
–175
250 500 750 1000
Figure 20. Low Power FFT, 1 MHz, −6 dB Input Tone, 8× Decimation
0
–200
0 1250
04975-022
FREQUENCY (kHz)
AMPLITUDE (dB)
–25
–50
–75
–100
–125
–150
–175
250 500 750 1000
TONE A: 999.75kHz
TONE B: 1.00025MHz
Figure 21. Low Power IMD, 1 MHz Center Frequency, 8× Decimation
0
–160
0 10000
04975-023
FREQUENCY (kHz)
AMPLITUDE (dB)
–20
–40
–60
–80
–100
–120
–140
2000 4000 6000 8000
TONE A: 999.75kHz
TONE B: 1.00025MHz
SECOND-ORDER IMD: –115.7dB
Figure 22. Low Power IMD, 1 MHz Center Frequency, 8× Decimation
AD7760
Rev. A | Page 15 of 36
0
–160
995 1005
04975-024
FREQUENCY (kHz)
AMPLITUDE (dB)
–20
–40
–60
–80
–100
–120
–140
997 999 1001 1003
TONE A: 999.75kHz
TONE B: 1.00025MHz
THIRD-ORDER
IMD: –89.15dB
Figure 23. Normal Mode IMD, 1 MHz Center Frequency, 8× Decimation
100.5
96.5
04
0
04975-025
MCLK FREQUENCY (MHz)
SNR (dBFS)
100.0
99.5
99.0
98.5
98.0
97.5
97.0
10 20 30
LOW POWER MODE
NORMAL MODE
Figure 24. SNR vs. MCLK Frequency, 8× Decimation, −6 dB, 1 kHz Input Tone
120
95
0 256
04975-026
DECIMATION RATE
SNR (dBFS)
115
110
105
100
64 128 192
–0.5dB
–6dB
–60dB
Figure 25. Normal Mode SNR vs. Decimation Rate, 1 kHz Input Tone
0
–160
995 1005
04975-027
FREQUENCY (kHz)
AMPLITUDE (dB)
–20
–40
–60
–80
–100
–120
–140
997 999 1001 1003
TONE A: 999.75kHz
TONE B: 1.00025MHz
THIRD-ORDER
IMD: –87.67dB
Figure 26. Normal Mode IMD, 1 MHz Center Frequency, 8× Decimation
–
105
–117
040
04975-028
MCLK FREQUENCY (MHz)
THD (dBc)
–107
–109
–111
–113
–115
10 20 30
NORMAL MODE
LOW POWER MODE
Figure 27. THD vs. MCLK Frequency, 8× Decimation, −6 dB, 1 kHz Input Tone
116
96
0 256
04975-029
DECIMATION RATE
SNR (dBFS)
112
108
104
100
64 128 192
–0.5dB
–6dB
–60dB
Figure 28. Low Power SNR vs. Decimation Rate, 1 kHz Input Tone
AD7760
Rev. A | Page 16 of 36
4500
0
8385222
04975-030
24-BIT CODE
OCCURRENCE
4000
3500
3000
2500
2000
1500
1000
500
8385238 8385254 8385270
Figure 29. Normal Mode, 24-Bit Histogram, 256× Decimation
600
0
8385016
04975-031
24-BIT CODE
OCCURRENCE
500
400
300
200
100
8385116 8385216 83854168385316 8385516
Figure 30. Normal Mode, 24-Bit Histogram, 8× Decimation
0.0010
–0.0010
0 1677721
6
04975-034
24-BIT CODE
INL (%)
0.0005
0
–0.0005
4194304 8388608 12582912
+25°C
–40°C
+85°C
Figure 31. 24-Bit INL, Normal Mode
3000
0
8383530
04975-032
24-BIT CODE
OCCURRENCE
2500
2000
1500
1000
500
83835246 8383562 8383578 8383594 8383610
Figure 32. Low Power, 24-Bit Histogram, 256× Decimation
450
0
8383236
04975-033
24-BIT CODE
OCCURRENCE
8383386 8383536 8383686 8383836
400
350
300
250
200
150
100
50
Figure 33. Low Power, 24-Bit Histogram, 8× Decimation
0.0015
–0.0010
0 1677721
6
04975-036
24-BIT CODE
INL (%)
0.0005
0.0010
0
–0.0005
4194304 8388608 12582912
+25°C
–40°C
+85°C
Figure 34. 24-Bit INL, Low Power Mode
AD7760
Rev. A | Page 17 of 36
0.6
–0.6
0 1677721
6
04975-035
24-BIT CODE
DNL (LSB)
4194304 8388608 12582912
0.4
0.2
0
–0.2
–0.4
Figure 35. 24-Bit DNL
04975-057
020
0
40
ICLK FREQUENCY (MHz)
AI
DD
2 (mA)
35
30
25
20
15
10
5
2 4 6 8 10 12 14 16 18
0
Figure 36. AIDD2 vs. ICLK Frequency (AVDD2 = 5 V)
04975-058
02
0
60
ICLK FREQUENCY (MHz)
DI
DD
(mA)
2 4 6 8 10 12 14 16 18
50
40
30
20
10
Figure 37. Decimate × 8, DIDD vs. ICLK Frequency (DVDD = 2.5 V)
04975-059
02
0
50
ICLK FREQUENCY (MHz)
DI
DD
(mA)
02 4 6 8 10 12 14 16 18
45
40
35
30
25
20
15
10
5
Figure 38. Decimate × 32, DIDD vs. ICLK Frequency (DVDD = 2.5 V)
04975-060
020
0
45
ICLK FREQUENCY (MHz)
DI
DD
(mA)
40
35
30
25
20
15
10
5
2 4 6 8 10 12 14 16 18
Figure 39. Decimate × 256, DIDD vs. ICLK Frequency (DVDD = 2.5 V)
AD7760
Rev. A | Page 18 of 36
THEORY OF OPERATION
The AD7760 employs a Σ- 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.
By employing oversampling, the quantization noise is spread
across a wide bandwidth from 0 to fICLK. This means that the
noise energy contained in the signal band of interest is reduced
(see Figure 40a). To further reduce the quantization noise in the
signal band of interest, a high order modulator is employed to
shape the noise spectrum so that most of the noise energy is
shifted out of the signal band (see Figure 40b).
The digital filtering that follows the modulator removes the
large out-of-band quantization noise (see Figure 40c) 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 in terms of phase.
The AD7760 employs three FIR filters in series. By using
different combinations of decimation ratios, filter selection,
and bypassing, data can be obtained from the AD7760 at a large
range of data rates. Multibit data from the modulator can be
obtained at the ICLK rate (see Modulator Data Output Mode
section). The first filter receives the data from the modulator at
a maximum frequency of 20 MHz and decimates it by 4 to output
the data at 5 MHz. The partially filtered data can be output at
this stage. The second filter allows the decimation rate to be
chosen from 2× to 32× or to be completely bypassed.
The third filter has a fixed decimation rate of 2×, is user
programmable, and has a default configuration. It is described
in detail in the Programmable FIR Filter section. This filter can
also be bypassed.
Table 6 shows some characteristics of the default filter. The
group delay of the filter is defined to be the delay to the center
of the impulse response and is equal to the computation plus
the filter delays. The delay until valid data is available (the
DVALID status bit is set) is equal to twice the filter delay plus
the computation delay.
QUANTIZATION NOISE
f
ICLK
\2
BAND OF INTEREST
a.
f
ICLK
\2
NOISE SHAPING
BAND OF INTEREST
b.
f
ICLK
\2
BAND OF INTEREST
DIGITAL FILTER CUTOFF FREQUENCY
c.
04975-037
Figure 40. Σ-Δ ADC
Table 6. Configuration with Default Filter
ICLK
Frequency Filter 1 Filter 2 Filter 3 Data State
Computation
Delay Filter Delay
Pass-Band
Bandwidth
Output Data
Rate (ODR)
20 MHz Bypassed Bypassed Bypassed Unfiltered 0 0 10 MHz 20 MHz
20 MHz 4× Bypassed Bypassed Partially filtered 0.325 µs 1.2 µs 1.35 MHz 5 MHz
20 MHz 4× Bypassed 2× Fully filtered 1.075 µs 10.8 µs 1 MHz 2.5 MHz
20 MHz 4× 2× Bypassed Partially filtered 1.35 µs 3.6 µs 562.5 kHz 2.5 MHz
20 MHz 4× 2× 2× Fully filtered 1.625 µs 22.8 µs 500 kHz 1.25 MHz
20 MHz 4× 4× Bypassed Partially filtered 1.725 µs 6 µs 281.25 kHz 1.25 MHz
20 MHz 4× 4× 2× Fully filtered 1.775 µs 44.4 µs 250 kHz 625 kHz
20 MHz 4× 8x Bypassed Partially filtered 2.6 µs 10.8 µs 140.625 kHz 625 kHz
20 MHz 4× 8× 2× Fully filtered 2.25 µs 87.6 µs 125 kHz 312.5 kHz
20 MHz 4× 16× Bypassed Partially filtered 4.175 µs 20.4 µs 70.3125 kHz 312.5 kHz
20 MHz 4× 16× 2× Fully filtered 3.1 µs 174 µs 62.5 kHz 156.25 kHz
20 MHz 4× 32× Bypassed Partially filtered 7.325 µs 39.6 µs 35.156 kHz 156.25 kHz
20 MHz 4× 32× 2× Fully filtered 4.65 µs 346.8 µs 31.25 kHz 78.125 kHz
12.288 MHz 4× 8× 2× Fully filtered 3.66 µs 142.6 µs 76.8 kHz 192 kHz
12.288 MHz 4× 16× 2× Fully filtered 5.05 µs 283.2 µs 38.4 kHz 96 kHz
12.288 MHz 4× 32× Bypassed Partially filtered 11.92 µs 64.45 µs 21.6 kHz 96 kHz
12.288 MHz 4× 32× 2× Fully filtered 7.57 µs 564.5 µs 19.2 kHz 48 kHz
AD7760
Rev. A | Page 19 of 36
MODULATOR DATA OUTPUT MODE
Operating the AD7760 in modulator output mode enables the
output of data directly from the Σ- modulator. This mode of
operation bypasses the AD7760 on-board digital filtering
capabilities, outputting data in its unfiltered form.
As discussed in the Theory of Operation section, the AD7760
operates using oversampling, which spreads quantization noise
over a wide bandwidth. The decrease in the quantization noise
energy in the resulting signal band is illustrated in Figure 40a.
By coupling the use of oversampling with the use of a high
order, multibit Σ- modulator, the AD7760 further reduces the
quantization noise in the signal band. Figure 41 is an FFT of
unfiltered data output from the AD7760 when it is used in
modulator output mode. This clearly demonstrates the shaping
of the quantization noise performed by the AD7760’s Σ-
modulator.
MODULATOR INPUTS
The maximum voltage input to each differential modulator
input pin is 0.8 × 4.096 V ≈ 3.275 V (80% of VREF), which must
sit on a common mode of VREF/2. This maximum differential
input voltage is shown as the conditioned output of the AD7760’s
on-board differential amplifier in Figure 52 in the Driving the
AD7760 section.
Further details on the signal conditioning implemented by the
AD7760’s on-board differential amplifier and the recommended
external circuitry that accompanies it is described in the
Driving the AD7760 section.
0
–160
01
FREQUENCY (MHz)
AMPLITUDE (dB)
0
–10
–20
–30
–40
–50
–60
–70
–80
–90
–100
–110
–120
–130
–140
–150
123456789
04975-048
Figure 41. FFT of Data Output by the AD7760 in Modulator Output Mode
MODULATOR DATA OUTPUT SCALING
In modulator output mode, data is output in a 16-bit twos
complement format on Pins D [15:0]; however, this data is
scaled to 15 bits. The transfer function in Figure 42 shows the
scaling involved for the 16 data bits output from Modulator
Pins D[15:0] vs. the maximum differential voltage input
allowed for the modulator inputs (VIN+ and VIN−).
D[15:0]
0011 0011 0011 0010
0011 1111 1111 1111
0000 0000 0000 0000
1100 1100 1100 1100
1100 0000 0000 0000
+4.096V
–4.096V
V
IN
+ = 3.6855V
V
IN
– = 0.4105V
+3.275V = MODULATOR FULL SCALE = 80% OF +4.096V
80% OF +4.096V = MODULATOR FULL SCALE = –3.275V
V
IN
+=2.048V
V
IN
– = 2.048V
V
IN
+ = 0.4105V
V
IN
– = 3.6855V
04975-049
Figure 42. Modulator Output Data Scaling
As the nature of the modulator output is coarse relative to
the fully filtered output of the AD7760 (due to the associated
quantization noise of the modulator output), Bits D[3:0] of the
modulator output are zero when operating in modulator data
output mode. Thus, the data outputs for the calculations listed
in Example 1 and Example 2 for inputs to the modulator pins
VIN+ and VIN− show Bits D[3:0] of the modulator output as zero.
Example 1
VIN+ = 3.5 V
VIN− = 0.595 V
Modulator Output Code = ([VIN(+) − VIN(−)]/4.096 V) × 16384
= [(3.5 V − 0.595 V)/4.096 V] × 16384
= +11620
Direct Scaling: [0010 1101 0110 0100]
Value Output on Data Output Pins D[15:0]:
D [15:0] = [0010 1101 0110 0000].
Example 2
VIN+ = 0.595 V
VIN− = 3.5 V
Modulator Output Code = ([VIN(+) − VIN(−)]/4.096 V) × 16384
= [(0.595 V − 3.5 V)/4.096 V] × 16384
= −11620
Direct Scaling: [1101 0010 1001 1100]
Value Output on Data Output Pins D[15:0]:
D [15:0] = [1101 0010 1001 0000].
AD7760
Rev. A | Page 20 of 36
MODULATOR DATA OUTPUT MODE INTERFACE
The AD7760 can be configured in modulator data output mode
(bypassing the default decimation filtering) by writing 0 to each
of the bits contained in Control Register 1: BYP F1, BYP F3,
and DEC [2:0]. This will bypass all digital decimation filtering
offered by the AD7760. See the AD7760 Registers section for
further details.
When the AD7760 is operating in modulator data output mode,
a different parallel interfacing scheme than that used for config-
urations, where the AD7760’s data output is filtered is necessary.
The data output rate depends on the clock divider ratio that is
used. When the CDIV bit in Control Register 2 is set to logic
high, data is output at the MCLK frequency. If the CDIV bit is
set to logic low, data is output at a frequency of MCLK/2. See
the Clocking the AD7760 section.
CLOCK DIVIDE-BY-1 MODE (CDIV = 1)
When obtaining data from the AD7760 in modulator output
mode, both the RD/WR and CS lines must be held low. This
brings the data bus out of its high impedance state. Figure 43
shows the timing diagram for reading data in the modulator data
output mode when operating with CDIV = 1 (that is, ICLK =
MCLK). A DRDY pulse is generated for each word. The data on
each of the 16 data output pins, D [15:0], is valid on the rising
edge of the DRDY pulse. The DRDY pulse can be used to latch
the modulator data into a FIFO or as a DMA control signal. Shortly
after the RD/WR and CS lines return high, the AD7760 stops
outputting data and the data bus returns to high impedance.
CLOCK DIVIDE-BY-2 MODE (CDIV = 0)
When operating in modulator output mode with CDIV = 0
(that is, ICLK = MCLK/2), the frequency of the DRDY signal
created is half that of the MCLK frequency input to the device.
The timing scheme that is used when CDIV = 0 depends on the
number of MCLK cycles that occur between RESET and SYNC.
If the number of MCLK cycles (n) between the rising edge of
RESET and the rising edge of SYNC (see Figure 44) is an even
value, use the interface timing shown in Figure 43. If n is an odd
value, use the interface timing shown in Figure 45.
t
9
t
10
t
14
t
11
t
12
t
13
DRDY
CS, RD/WR
D[0:15] INVALID DATA MOD DATA M MOD DATA M + 1 MOD D...
04975-050
Figure 43. AD7760 Modulator Output Mode (CDIV = 1) and (CDIV = 0, n is even)
MCLK
RESET
SYNC
n×t
MCLK
04975-051
Figure 44. AD7760 Relative Timing Between RESET and SYNC in Modulator Output Mode CDIV = 0
AD7760
Rev. A | Page 21 of 36
INVALID DATA MOD DATA M MOD DATA M + 1 MOD D...
t9t10
t20
DRDY
D[0:15]
MCLK
CS, RD/WR
t11
t19
t14
04975-052
Figure 45. AD7760 Modulator Output Mode (CDIV = 0, n is odd)
In the case where n is an odd number of MCLK cycles, the
modulator data output on Pins D [15:0] is output on the rising
edge of DRDY. In this case, the modulator data should be read
on the falling edge of MCLK when DRDY is logic low. Figure 45
shows timing details to be used when reading the modulator
output data where CDIV = 0 and there is an odd number of
MCLK cycles between the rising edge of RESET and the rising
edge of SYNC. The edge of MCLK that should be used under
these conditions is illustrated in Figure 45 by arrows on the
MCLK falling edges in question.
USING THE AD7760
IN MODULATOR OUTPUT MODE
The following is the recommended sequence for powering up
and using the AD7760:
1. Apply power.
2. Start the clock oscillator, applying MCLK.
3. Ta ke RESET low for a minimum of one MCLK cycle.
4. Wait a minimum of two MCLK cycles after the rising edge
of RESET.
5. Write to Control Register 2 to power up the ADC and the
differential amplifier as required. The correct clock divider
(CDIV) ratio should be programmed at this time.
6. Write to Control Register 1 to set the bypass filter bits, BYP F1
and BYP F3, and the decimation rate bits, DEC [2:0], to 0.
7. Wait a minimum of six MCLK cycles after the rising edge
of CS has been released.
8. Ta ke SYNC low for a minimum of four MCLK cycles, if
required, to synchronize multiple parts.
Using this sequence results in an even number of MCLK cycles
between the rising edge of RESET and the rising edge of SYNC.
Therefore, when using this sequence with CDIV = 0, the interface
timing shown in Figure 43 should be implemented.
Note that whether the number of MCLK cycles between the
rising edge of RESET and SYNC is odd or even is irrelevant
when the AD7760 is operated with CDIV = 1.
When using the AD7760 in modulator output mode, the offset,
gain, and overrange registers are not operational. The only
registers that can be used are Control Register 1 and Control
Register 2.
AD7760
Rev. A | Page 22 of 36
AD7760 INTERFACE
READING DATA
When the AD7760 is outputting data at a 5 MHz output data
rate or less, the interface operates in a conventional mode, as
shown in Figure 2, using a 16-bit bidirectional parallel interface.
This interface is controlled by the RD/WR and CS pins. The 24-bit
conversion data is output in twos complement format. When a
new conversion result is available, an active low pulse is output
on the DRDY pin.
To read a conversion result from the AD7760, two 16-bit read
operations are performed. The DRDY pulse indicates that a new
conversion result is available. Both RD/WR and CS go low to
perform the first read operation. Shortly after both lines go low,
the data bus becomes active and the 16 most significant bits
(MSBs) of the conversion result are output. The RD/WR and CS
lines must return high for a full ICLK period before the second
read is performed. This second read contains the eight least
significant bits (LSBs) of the conversion result along with six
status bits. These status bits are shown in Table 7. Descriptions
of the other status bits are found in Table 17.
Table 7. Status Bits During Data Read
MSB LSB
D7 D6 D5 D4 D3 D2 D1 D0
DVALID OVR UFILT LPWR FILTOK DLOK 0 0
Shortly after RD/WR and CS return high, the data bus returns
to a high impedance state. Both read operations must be
completed before a new conversion result is available because
the new result overwrites the contents of the output register.
If a DRDY pulse occurs during a read operation, the data read
is invalid.
READING STATUS AND OTHER REGISTERS
The AD7760 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 that corresponds to the register to be read. The next
read operation outputs the contents of the selected register
instead of a conversion result. The AD7760 Registers section
provides more information on the relevant bits in the control
register.
SHARING THE PARALLEL BUS
By its nature, the high accuracy of the AD7760 makes it
sensitive to external noise sources. These include digital activity
on the parallel bus. For this reason, it is recommended that the
AD7760 data lines be isolated from the system data bus by
means of a latch or buffer to ensure all digital activity on the
D0 to D15 pins is controlled by the AD7760.
If multiple synchronized AD7760 parts that share a properly
distributed common MCLK signal exist in a system, these parts
can share a common bus without being isolated from each
other. This bus can then be isolated from the system bus by a
single latch or buffer.
SYNCHRONIZATION
The SYNC input to the AD7760 provides a synchronization
function that allows the user to begin gathering samples of the
analog front-end input from a known point in time.
The SYNC function allows multiple AD7760s, operated from
the same MCLK, RESET, and SYNC signals, to be synchronized
so that each ADC simultaneously updates its output register.
The distribution of the signals that are common to each of the
devices that are to be synchronized is extremely important in
ensuring that the timing of each of the AD7760 devices is
correct, that is, that each AD7760 device sees the same digital
edges synchronously.
The SYNC signal is sensed on the falling edge of MCLK. On the
first falling edge of MCLK after SYNC goes logic low, the digital
filter sequencer is reset to 0. The filter is held in a reset state until
a falling edge of the MCLK senses SYNC logic high. The SYNC
signal must remain logic low for a minimum of four MCLK
cycles. Figure 46 shows the recommended timing for the SYNC
signal with respect to MCLK.
DEVICE SYNCHRONIZED
FROM THIS POINT IN TIME
MCLK
SYNC 4×
t
MCLK
MINSYNCLOGICLOW
04975-053
Figure 46. Recommended SYNC Timing
The rising edge of SYNC should be coincident with the rising
edge of MCLK. Thus, the next falling edge of MCLK senses
SYNC logic high and takes the filter out of its reset state. By
applying this signal scheme to multiple ADCs using the same
MCLK and SYNC signals, all of the devices will gather input
samples synchronously.
Following a SYNC signal, the digital filter needs time to settle
before valid data can be read from the AD7760. The DVALID
status bit (D7 in Table 7) output with each conversion indicates
when valid data is being output by the converter. The time from
the rising edge of SYNC until the DVALID bit is asserted is
dependent on the filter configuration used. See the Theory of
Operation section and the values listed in Table 6 for details on
calculating the time until DVALID is asserted.
AD7760
Rev. A | Page 23 of 36
WRITING TO THE AD7760
There are many features and parameters that the user can
change by writing to the AD7760 device. See the Using the
AD7760 section, which details the writing sequence needed to
initialize the operation of the part.
The AD7760 has programmable registers that are 16 bits wide.
This means that two write operations are required to program a
register. The first write contains the register address, and the
second write contains the register data. An exception is when a
user-defined filter is being downloaded to the AD7760. This is
described in detail in the Downloading a User-Defined Filter
section. The AD7760 Registers section contains the register
addresses and details.
Figure 3 shows a write operation to the AD7760. The RD/WR
line is held high while the CS line is brought low for a minimum
of four ICLK periods. The register address is latched during this
period. The CS line is brought high again for a minimum of
four ICLK periods before the register data is put onto the data
bus. If a read operation occurs between the writing of the
register address and the register data, the register address is
cleared and the next write must be the register address. This
also provides a method to revert back to a known situation if
the user forgets whether the next write is an address or data.
Generally, the AD7760 is written to and configured on power-
up and very infrequently, if at all, after that. Following any write
operation, the full group delay of the filter must elapse before
valid data is output from the AD7760.
AD7760
Rev. A | Page 24 of 36
CLOCKING THE AD7760
The AD7760 requires an external low jitter clock source. This
signal is applied to the MCLK pin, and the MCLKGND pin is
used to sense the ground from the clock source. An internal
clock signal (ICLK) is derived from the MCLK input signal. The
ICLK controls all internal operations of the AD7760. The
maximum ICLK frequency is 20 MHz, but due to an internal
clock divider, a range of MCLK frequencies can be used. There
are two ways to generate the ICLK:
ICLK = MCLK (CDIV = 1)
ICLK = MCLK/2 (CDIV = 0)
These options are selected from the control register (see the
AD7760 Registers section for more details). On power-up, the
default is ICLK = MCLK/2 to ensure that the part can handle
the maximum MCLK frequency of 40 MHz. For output data
rates equal to those used in audio systems, a 12.288 MHz ICLK
frequency can be used. As shown in Table 6, output data rates of
192 kHz, 96 kHz, and 48 kHz are achievable with this ICLK
frequency. As mentioned previously, this ICLK frequency can
be derived from different MCLK frequencies.
It is recommended that the MCLK signal applied to the
AD7760 has a 50-50 mark-space ratio. When operating in clock
divide-by-1 mode (that is, CDIV = 1), using higher mark-space
ratios reduces the maximum MCLK frequency that can be
applied to the AD7760 yielding maximum performance. For
example, using a mark-space ratio of 60-40 (with CDIV = 1)
reduces the maximum MCLK frequency that will yield the
maximum INL and THD performance to 16 MHz.
BUFFERING THE MCLK SIGNAL
The MCLK signal for the AD7760 must be buffered before
being input to the MCLK pin on the AD7760 device. This can
be done simply by routing the MCLK signal to both inputs of an
AND gate (see Figure 47).
The recommended buffer is the NC7SZ08M5, which is a two-
input AND gate from Fairchild Semiconductor. Using the buffer
with a supply voltage of 5 V is advised to achieve optimum
performance from the AD7760.
3
MCLK
MCLK
SOURCE
NC7SZ08M5
(AND GATE)
AD7760
04975-054
Figure 47. Buffering the MCLK Signal Using the NC7SZ08M5 AND Gate
MCLK JITTER REQUIREMENTS
The MCLK jitter requirements depend on a number of factors
and are given by
20
)(
102
)( dBSNR
f
OSR
t
IN
rmsj
××π×
=
where:
OSR = oversampling ratio = fICLK/ODR.
fIN = maximum input frequency.
SNR(dB) = target SNR.
Example 1
This example can be taken from Table 6, where:
ODR = 2.5 MHz.
fICLK = 20 MHz.
fIN (max) = 1 MHz.
SNR = 108 dB.
ps79.1
10102
8
4.56
)( =
××π×
=
rmsj
t
This is the maximum allowable clock jitter for a full-scale,
1 MHz input tone with the given ICLK and output data rate.
Example 2
Take a second example from Table 6, where:
ODR = 48 kHz.
fICLK = 12.288 MHz.
fIN (max) = 19.2 kHz.
SNR = 120 dB.
ps133
10102.192
256
63
)( =
×××π×
=
rmsj
t
The input amplitude also has an effect on these jitter figures.
For example, if the input level was 3 dB below full-scale, the
allowable jitter would be increased by a factor of √2, increasing
the first example to 2.53 ps rms. This happens when the
maximum slew rate is decreased by a reduction in amplitude.
Figure 48 and Figure 49 illustrate this point, showing the
maximum slew rate of a sine wave of the same frequency but
with different amplitudes.
AD7760
Rev. A | Page 26 of 36
DRIVING THE AD7760
The AD7760 has an on-chip differential amplifier that operates
with a supply voltage (AVDD3) within the 3.15 V to 5.25 V range.
For a 4.096 V reference, the supply voltage must be 5 V.
To achieve the specified performance in normal mode, the
differential amplifier should be configured as a first-order
antialias filter, as shown in Figure 50. 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 values in the table as an example yields a
10 dB attenuation at the first alias point of 19 MHz.
A1
R
IN
R
FB
C
FB
R
IN
R
M
R
M
C
S
R
FB
C
FB
V
IN
–
A
BV
IN
+
0
4975-040
Figure 50. Differential Amplifier Configuration
Table 8. Normal Mode Component Values
VREF R
IN R
FB R
M C
S C
FB
4.096 V 1 kΩ 655 Ω 18 Ω 5.6 pF 33 pF
Figure 52 shows the signal conditioning that occurs using the
circuit shown in Figure 50 with a ±2.5 V input signal biased
around ground and the component values and conditions listed
in Table 8. The differential amplifier always biases the output
signal to sit on the optimum common mode of VREF/2, in this
case 2.048 V. The signal is also scaled to give the maximum
allowable voltage swing with this reference value. This is
calculated as 80% of VREF, that is, 0.8 × 4.096 V ≈ 3.275 V p-p
on each input.
With a 4.096 V reference, a 5 V supply must be provided to the
reference buffer (AVDD4). With a 2.5 V reference, a 3.3 V supply
must be provided to AVDD4.
Figure 51 shows the transfer function in terms of the 24-bit
digital output codes (twos complement coding) of the AD7760
vs. the voltage signals VA and VB applied to the on-board
differential amplifier A1, as shown in Figure 52.
011…111
011…110
000…010
000…001
000…000
111…111
111…110
100…000
100…001
24 BITS
AD7760
24-BIT
O
UTPUT
B=+2.5V
A = –2.5V
A=0V
B=0V
A=+2.5V
B = –2.5V
04975-056
Figure 51. Transfer Function for the AD7760 Filtered Output Where VA and VB
are Inputs to the On-Board Differential Amplifier A1
+2.5V
0V
–2.5V
A
+2.5V
0V
–2.5V
B
INPUTS TO THE AD7760 DIFFERENTIAL AMPLIFIER OUTPUTS OF THE AD7760 DIFFERENTIAL AMPLIFIER
+3.685V
+2.048V
+0.410V
+3.685V
+2.048V
+0.410V
V
IN
+
V
IN
–
0
4975-055
Figure 52. Differential Amplifier Signal Conditioning
AD7760
Rev. A | Page 27 of 36
To obtain maximum performance from the AD7760, it is
advisable to drive the ADC with differential signals. Figure 53
shows how a bipolar, single-ended signal biased around ground
can drive the AD7760 with the use of an external op amp, such
as the AD8021.
A1
R
IN
R
FB
C
FB
R
IN
R
M
R
M
C
S
R
FB
C
FB
V
IN
–
V
IN
V
IN
+
AD8021
2R
2R
R
0
4975-042
Figure 53. Single-Ended-to-Differential Conversion
The AD7760 employs a double-sampling front end, as shown in
Figure 54. For simplicity, only the equivalent input circuit for VIN+
is shown. The equivalent input circuitry for VIN− is the same.
CS2
CPB2 SS4
SH4
CPA
SS2
SH2
CS1
CPB1 SS3
SH3
SS1
SH1
ANALOG
MODULATOR
V
IN
+
04975-043
Figure 54. Equivalent Input Circuit
Sampling Switches SS1 and SS3 are driven by ICLK, whereas
Sampling Switches SS2 and SS4 are driven by ICLK. When ICLK is
high, the analog input voltage is connected to CS1. On the falling
edge of ICLK, the SS1 and SS3 switches open and the analog input
is sampled on CS1. Similarly, when ICLK is low, the analog input
voltage is connected to CS2. On the rising edge of ICLK, the SS2
and SS4 switches open and the analog input is sampled
on CS2.
Capacitors CPA, CPB1, and CPB2 represent parasitic capacitances
that include the junction capacitances associated with the MOS
switches.
Table 9. Equivalent Component Values
Mode CS1 (pF) CS2 (pF) CPA (pF) CPB1/2 (pF)
Normal 51 51 12 20
Low Power 13 13 12 5
USING THE AD7760
The following is the recommended sequence for powering up
and using the AD7760:
1. Apply power.
2. Start the clock oscillator, applying MCLK.
3. Ta ke RESET low for a minimum of one MCLK cycle.
4. Wait a minimum of two MCLK cycles after RESET has
been released.
5. Write to Control Register 2 to power up the ADC and the
differential amplifier as required. The correct clock divider
(CDIV) ratio should be programmed at this time.
6. Write to Control Register 1 to set the output data rate.
7. Wait a minimum of five MCLK cycles after CS has been
released.
8. Ta ke SYNC low for a minimum of four MCLK cycles, if
required, to synchronize multiple parts.
Data can then be read from the part using the default filter,
offset, gain, and overrange threshold values. The conversion
data read is not valid, however, until the group delay of the filter
has elapsed. Once this has occurred, the DVALID bit read with
the data LSW is set, indicating that the data is indeed valid.
The user can then download a different filter if required (see the
Downloading a User-Defined Filter section). Values for gain,
offset, and overrange threshold registers can be written or read
at this stage.
AD7760
Rev. A | Page 28 of 36
DECOUPLING AND LAYOUT RECOMMENDATIONS
Due to the high performance nature of the AD7760, correct decoupling and layout techniques are required to obtain the performance as
stated within this data sheet. Figure 55 shows a simplified connection diagram for the AD7760.
VINA+
VINA–
VOUTA–
VOUTA+
VINA+
VINA–
VOUTA–
VOUTA+
DECAPA
DECAPB
VIN+
VIN–
VREF+
VIN+
VIN–
VREFX
REFGND
RBIAS
DGND
DGND
DGND
DGND
DGND
DGND
DGND
19
20
21
22
8
30
25
26
10
9
17
1
35
42
43
53
62
64
DB0
DB2
DB1
DB3
DB4
DB5
DB6
DB7
DB8
DB9
DB10
DB11
DB12
DB13
DB14
DB15
DB0
DB2
DB1
DB3
DB4
DB5
DB6
DB7
DB8
DB9
DB10
DB11
DB12
DB13
DB14
DB15
CS
RD/WR
RESET
SYNC
DRDY
MCLK
MCLKGND
CS
RD/WR
RESET
SYNC
DRDY
MCLK
AGND1
AGND1
AGND2
AGND2
AGND2
AGND2
AGND2
AGND3
AGND3
AGND3
AGND3
AGND4
7
34
5
13
16
18
28
23
29
31
32
11
AVDD2
AVDD2
AVDD2
AVDD4
AVDD1
AVDD1
AVDD3
AVDD2
VDRIVE
VDRIVE
DVDD
PIN 1
4
PIN 1
5
PIN 4
PIN 1
2
PIN 6
PIN 3
3
PIN 2
4
PIN 2
7
PIN 4
4
PIN 6
3
PIN 41
14
15
4
12
6
33
24
27
44
63
41
AD7760BSV
61
60
59
58
55
54
50
49
46
45
40
37
36
38
3
2
57
56
52
51
48
47
39
R19
160kΩ
C64
33pF
C7
100nF
DB [0:15]
U2
AVDD3
PIN 24
(VDIF1)
C54
100nF
L6
DVDD
PIN 41
(DVDD)
C58
100nF
L8
AVDD2
PIN 4
(RHS)
C48
100nF
L1
PIN 15
(VBIAS)
C50
100nF
L3
PIN 14
(LHS) PIN 27
C62
100nF
L2
L9
AVDD4
PIN 12
(VBUF)
C59
10nF
L4
R38
10Ω
AVDD1
PIN 5
(VMOD1)
C52
100nF
L5
PIN 33
(VMOD2)
C53
100nF
L11
VDRIVE
PIN 44
(VDRV1)
C56
100nF
L7
PIN 63
(VDRV2)
C57
100nF
L12
04975-046
Figure 55. Simplified Connection Diagram
AD7760
Rev. A | Page 29 of 36
SUPPLY DECOUPLING
Every supply pin must be connected to the appropriate supply
via a ferrite bead and decoupled to the correct ground pin with
a 100 nF, 0603 case size, X7R dielectric capacitor. There are two
exceptions to this:
• Pin 12 (AVDD4) must have a 10 resistor inserted between
the pin and a 10 nF decoupling capacitor, which is
connected to ground at Pin 9.
• Pin 27 (AVDD2) does not require a separate decoupling
capacitor or a direct connection to the supply, but instead
is connected to Pin 14 via a 15 nH inductor.
ADDITIONAL DECOUPLING
There are two other decoupling pins on the AD7760—Pin 8
(DECAPA) and Pin 30 (DECAPB). Pin 8 should be decoupled
with a 100 nF capacitor, and Pin 30 requires a 33 pF capacitor.
REFERENCE VOLTAGE FILTERING
A low noise reference source, such as the ADR431 (2.5 V) or
ADR434 (4.096 V), is suitable for use with the AD7760. The
reference voltage supplied to the AD7760 should be decoupled
and filtered as shown in Figure 56.
The recommended scheme for the reference voltage supply is a
100 series resistor connected to a 100 F tantalum capacitor,
followed by a series resistor of 10 , and finally a 10 nF capacitor
placed as close as possible to the VREF+ pin, decoupling this
capacitor to the associated ground pin, Pin 11.
7.5V PIN 10V
OUT
2V
IN
6
4
C15
10µF
C9
100nF
R30
100Ω
R17
10Ω
+C46
10nF
C11
100µF
+
ADR434
GND
U3
04975-047
Figure 56. Reference Connection
DIFFERENTIAL AMPLIFIER COMPONENTS
The correct components for use around the on-chip differential
amplifier are detailed in Table 8. Matching the components on
both sides of the differential amplifier is important to minimize
distortion of the signal applied to the amplifier. A tolerance of
0.1% or better is required for these components. Symmetrical
routing of the tracks on both sides of the differential amplifier
also assists in achieving the stated performance.
BIAS RESISTOR SELECTION
The AD7760 requires a resistor to be connected between the
RBIAS and AGND pins. The value of this resistor is dependent
on the reference voltage being applied to the device. The resistor
value should be selected to produce a current of 25 µA through
the resistor to ground. For a 2.5 V reference voltage, the correct
resistor value is 100 kΩ, and for a 4.096 V reference, the correct
resistor value is 160 kΩ.
LAYOUT CONSIDERATIONS
While using the correct components is essential to achieve
optimum performance, the correct layout is just as important.
The AD7760 product page on the Analog Devices website
contains the Gerber files for the AD7760 evaluation board.
These files should be used as a reference when designing any
system using the AD7760.
The location and orientation of some of the components
mentioned in previous sections of this data sheet are critical,
and particular attention must be paid to the components that
are located close to the AD7760. Locating these components
farther away from the device can have a direct impact on the
achievable maximum performance.
The use of ground planes should also be carefully considered.
To ensure that the return currents through the decoupling
capacitors are flowing to the correct ground pin, the ground side
of the capacitors should be as close as possible to the ground pin
associated with that supply. A ground plane should not be relied
on as the sole return path for decoupling capacitors because the
return current path using ground planes is not easily predictable.
EXPOSED PADDLE
The AD7760 64-lead TQFP employs a 6 mm × 6 mm exposed
paddle (see Figure 59). The paddle reduces the thermal
resistance of the package by providing a path for heat energy to
flow between the package and the PCB and, in turn, increases
the heat transfer efficiency from the AD7760 package.
Connecting the exposed paddle to the AGND plane of the PCB
is essential in creating the conditions that allow the AD7760
package to perform to the highest specifications possible. The
exposed paddle should not be connected directly to any of the
ground pins on the AD7760 and should only be connected to
the analog ground plane. Best practice is to use multiple vias
connecting the exposed paddle to the AGND plane of the PCB.
AD7760
Rev. A | Page 30 of 36
PROGRAMMABLE FIR FILTER
As previously mentioned, the third FIR filter on the AD7760
is user programmable. The default coefficients that are loaded
upon reset are given in Table 10, and the frequency responses
are shown in Figure 57. The frequencies quoted in Figure 57
scale directly with the output data rate.
Table 10. Default Filter Coefficients
No.
Dec
Value
Hex
Value No.
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 most applications. It is
a standard brick wall filter with a symmetrical impulse response.
The default filter has a length of 96, is nonaliasing, and provides
120 dB of attenuation at Nyquist. This filter not only performs
signal antialiasing, but also suppresses out-of-band quantization
noise produced by the analog-to-digital conversion process. Any
significant relaxation in the stop-band attenuation or transition
bandwidth relative to the default filter can result in a failure to
meet the SNR specifications.
The default filter characteristics scale with both the MCLK
frequency applied and the decimation rate chosen by the user.
To create a filter, note the following:
• The filter must be an 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 taps and 96 taps in
steps of 12.
• Because the filter is symmetrical, the number of
coefficients that must be downloaded is half the filter
length. The default filter coefficients exemplify this with
only 48 coefficients listed for a 96-tap filter.
• Coefficients are written from the center of the 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 134,217,726, with the coefficients rounded
to the nearest integer. For a low-pass filter, this is the
equivalent of having the coefficients summed arithmetically
(including sign) to a +67,108,863 (0x3FF FFFF) positive
value over the half-impulse-response coefficient set (a
maximum of 48 coefficients). Any deviation from this
introduces a gain error.
–20
–40
–60
–80
–100
–120
–140
–160
0 500 200015001000 2500
FREQUENCY (kHz)
AMPLITUDE (dB)
0
–0.1dB FREQUENCY = 1.004MHz
PASS-BAND RIPPLE = 0.05dB
STOP BAND = 1.25MHz
–3dB FREQUENCY = 1.06MHz
04975-044
Figure 57. Default Filter Frequency Response (2.5 MHz ODR)
The procedure for downloading a user-defined filter is detailed
in the Downloading a User-Defined Filter section.
AD7760
Rev. A | Page 31 of 36
DOWNLOADING A USER-DEFINED FILTER
As previously mentioned, the filter coefficients are 27 bits in
length—one sign and 26 magnitude bits. Because the AD7760
has a 16-bit parallel bus, the coefficients are padded with 5 MSB
0s to generate a 32-bit word, split into two 16-bit words for
downloading. The first 16-bit word for each coefficient becomes
(00000, sign bit, Magnitude [25:16]), whereas the second word
becomes (Magnitude [15:0]). To ensure that a filter is downloaded
correctly, a checksum must also be generated and then downloaded
following the final coefficient. The checksum is a 16-bit word
generated by splitting each 32-bit word into four bytes and
summing the bytes from all coefficients up to a maximum of
192 bytes (48 coefficients × four bytes). The same checksum
is generated internally in the AD7760 and compared with the
downloaded checksum. The DL_OK bit in the status register is
set if these two checksums agree.
To download a user filter
1. Write to Control Register 1, setting the DL_FILT bit and
the correct filter length bits corresponding to the length of
the filter to be downloaded (see Table 11).
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 Step 2 and Step 3 for each coefficient.
5. Write the 16-bit checksum.
6. Use the following methods to verify that the filter coefficients
are downloaded correctly:
a. Read the status register, checking the DL_OK bit.
b. Read data and observe the status of the DL_OK bit.
Note that because the user coefficients are stored in RAM, they
are cleared after a RESET operation or a loss of power.
Table 11. Filter Length Values
FLEN [3:0] Number of Coefficients 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
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 58.
10
–80
06
FREQUENCY (kHz)
AMPLITUDE (dB)
00
0
–10
–20
–30
–40
–50
–60
–70
100 200 300 400 500
04975-045
Figure 58. 24-Tap FIR Frequency Response
The coefficients for the filter are listed in Table 12 and are
shown from the center of symmetry outwards. The raw
coefficients were generated using a commercial filter design
tool and were scaled appropriately so that their sum equals
67,108,863 (0x3FF FFFF).
Table 12. 24-Tap FIR Coefficients
Coefficient 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
AD7760
Rev. A | Page 32 of 36
Table 13 shows the hexadecimal values (in sign-and-magnitude
format) that are downloaded to the AD7760 to realize this filter.
The table is also split into the bytes that are summed to produce
the checksum. The checksum generated from these coefficients
is 0x0E6B.
Table 13. Filter Hexadecimal Values
Word 1 Word 2
Coefficient 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
Table 14 lists the 16-bit words the user would write to the
AD7760 to set up the ADC and download this filter, assuming
an output data rate of 1.25 MHz has been selected.
Table 14. Sequence of Write Instructions to Set Up Device
and Download the Filter Example
Word Description
0x0001 Address of Control Register 1.
0x8079 Control register data. DL filter: set filter length = 24,
set output data rate = 1.25 MHz.
0x032B First coefficient, Word 1.
0x9688 First coefficient, Word 2.
0x01BF Second coefficient, Word 1.
0x183C Second coefficient, Word 2.
… Other coefficients.
0x0404 Twelfth (final) coefficient, Word 1.
0x3B5F Final coefficient, Word 2.
0x0E6B Checksum. Wait (0.5 × tICLK × number of unused
coefficients) for AD7760 to write 0s to the remaining
unused coefficients.
0x0001 Address of control register.
0x0879 Control register data. Set read status and maintain
filter length and decimation settings. Read contents
of status register. Check Bit 7 (DL_OK) to determine
if the filter was downloaded correctly.
AD7760
Rev. A | Page 33 of 36
AD7760 REGISTERS
The AD7760 has a number of user-programmable registers. The control registers are used to set the decimation rate, the filter configuration, the
clock divider, and so on. There are also digital gain, offset, and overrange 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 in this section.
CONTROL REGISTER 1—ADDRESS 0x0001
Default Value 0x001A
MSB
LSB
D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0
DL_
FILT
RD
OVR
RD
GAIN
RD
OFF
RD
STAT
0
SYNC
FLEN3
FLEN2
FLEN1
FLEN0
BYP F3
BYP F1
DEC2
DEC1
DEC0
Table 15. Bit Descriptions of Control Register 1
Bit Mnemonic Description
15 DL_FILT1Download 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 are interpreted as the user coefficients for the FIR
filter until all the coefficients and the checksum have been written.
14 RD OVR1, 2Read Overrange. If this bit has been set, the next read operation outputs the contents of the overrange
threshold register instead of a conversion result.
13 RD GAIN1, 2 Read Gain. If this bit has been set, the next read operation outputs the contents of the digital gain register.
12 RD OFF1, 2 Read Offset. If this bit has been set, the next read operation outputs the contents of the digital offset register.
11 RD STAT1, 2 Read Status. If this bit has been set, the next read operation outputs the contents of the status register.
10 0 0 must be written to this bit.
9 SYNC1 Synchronize. Setting this bit initiates an internal synchronization routine. Setting this bit simultaneously on
multiple devices synchronizes all filters.
8 to 5 FLEN [3:0] Filter Length Bits. These bits must be set when the DL_FILT bit is set before a user-defined filter is downloaded.
4 BYP F3 Bypass Filter 3. If this bit is 0, Filter 3 (programmable FIR) is bypassed.
3 BYP F1 Bypass Filter 1. If this bit is 0, Filter 1 is bypassed. This should only occur when the user requires unfiltered
modulator data to be output.
2 to 0 DEC [2:0] Decimation Rate. These bits set the decimation rate of Filter 2. All 0s implies that the filter is bypassed. A value
of 1 corresponds to 2× decimation, a value of 2 corresponds to 4× decimation, and so on, up to the maximum
value of 5, corresponding to 32× decimation.
1 Bit 15 to Bit 9 are self-clearing bits.
2 Only one of the bits from Bit 14 to Bit 11 can be set in any write operation because it determines the contents of the next read operation.
CONTROL REGISTER 2—ADDRESS 0x0002
Default Value After RESET: 0x009B
Recommended register setting for power-up and normal operation using clock divide-by-2 (CDIV = 0) mode: 0x0002
MSB LSB
D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0
0
0
0
0
0
0
0
0
0
0 CDIV
0
PD
LPWR
1
D1PD
Table 16. Bit Descriptions of Control Register 2
Bit Mnemonic Description
5
CDIV
Clock Divider Bit. This sets the divide ratio of the MCLK signal to produce the internal ICLK. Setting CDIV = 0 divides the
MCLK by 2. If CDIV = 1, the ICLK frequency is equal to the MCLK.
3 PD Power Down. Setting this bit powers down the AD7760, reducing the power consumption to 6.35 mW.
2 LPWR Low Power. If this bit is set, the AD7760 is operating in a low power mode. The power consumption is reduced for a 6 dB
reduction in noise performance.
1 1 Write 1 to this bit.
0 D1PD Differential Amplifier Power Down. Setting this bit powers down the on-chip differential amplifier.
AD7760
Rev. A | Page 34 of 36
STATUS REGISTER (READ ONLY)
MSB LSB
D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0
PART 1
PART 0
DIE 2
DIE 1
DIE 0
0
LPWR
OVR
DL_OK
FILTOK
UFILT BYP F3 BYP F1
DEC2
DEC1
DEC0
Table 17. Bit Descriptions of Status Register
Bit Mnemonic Comment
15, 14 PART [1:0] Part Number. These bits are constant for the AD7760.
13 to 11 DIE [2:0] Die Number. These bits reflect the current AD7760 die number for identification purposes within a system.
10 0 This bit is set to 0.
9 LPWR Low Power. If the AD7760 is operating in low power mode, this bit is set to 1.
8 OVR If the current analog input exceeds the current overrange threshold, this bit is set.
7 DL_OK
When downloading a user filter to the AD7760, a checksum is generated. This checksum is compared to the one
downloaded following the coefficients. If these checksums agree, this bit is set.
6 FILTOK
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 UFILT 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 to 0 DEC [2:0] Decimation Rate. These bits correspond to the bits set in Control Register 1.
OFFSET REGISTER—ADDRESS 0x0003
Non-bit-mapped, Default Value 0x0000
The offset register uses twos complement notation and is scaled
such that 0x7FFF (maximum positive value) and 0x8000 (max-
imum 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 4.096 V, the offset correction range is
approximately ±25 mV.
GAIN REGISTER—ADDRESS 0x0004
Non-bit-mapped, 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 results in a full-scale digital output when the input is at
80% of VREF, tying in with the maximum analog input range of
±80% of VREF p-p.
OVERRANGE REGISTER—ADDRESS 0x0005
Non-bit-mapped, Default Value 0xCCCC
The overrange 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 = 4.096 V, the bit is then set when
the input voltage exceeds approximately 6.55 V p-p differential.
Once the overrange bit is set, the DVALID bit in the status bits
of the AD7760 ouptut is set to zero, providing another indication
that an input overrange has occurred. Note that the overrange
bit is set immediately if the analog input voltage exceeds 100% of
VREF for more than four consecutive samples at the modulator rate.
AD7760
Rev. A | Page 35 of 36
OUTLINE DIMENSIONS
COMPLIANT TO JEDEC STANDARDS MS-026-ACD-HD
1.05
1.00
0.95
0.20
0.09
0.08 MAX
COPLANARITY
VIEW A
ROTATED 90
°
CCW
SEATING
PLANE
0° MIN
7°
3.5°
0°
0.15
0.05
49
64
1
17
16
32
33
48
1.20
MAX
0.75
0.60
0.45
VIEW A
TOP VIEW
(PINS DOWN)
PIN 1
49 64
17
1
16
32
33
48
0.50
BSC
LEAD PITCH
0.38
0.32
0.22
BOTTOM VIEW
(PINS UP)
6.00
BSC SQ
EXPOSED
PAD
12.20
12.00 SQ
11.80
10.20
10.00 SQ
9.80
Figure 59. 64-Lead Thin Quad Flat Package, Exposed Pad [TQFP_EP]
(SV-64-2)
Dimensions shown in millimeters
ORDERING GUIDE
Model Temperature Range Package Description Package Option
AD7760BSVZ1−40°C to +85°C 64-Lead Thin Quad Flat Package, Exposed Pad [TQFP_EP] SV-64-2
AD7760BSVZ-REEL1−40°C to +85°C 64-Lead Thin Quad Flat Package, Exposed Pad [TQFP_EP] SV-64-2
EVAL-AD7760EB Evaluation Board
1 Z = Pb-free part.
AD7760
Rev. A | Page 36 of 36
NOTES
©2006 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D04975-0-8/06(A)
