AD AD7760BSV

2.5 MSPS, 20-Bit Σ∆ ADC
AD7760
Preliminary Technical Data
FEATURES
FUNCTIONAL BLOCK DIAGRAM
VIN- VIN+
Multi-Bit
Sigma-Delta
Modulator
DIFF
VREF+
+
BUF
-
Reconstruction
AD7760
Programmable
Decimation
Control Logic,
I/O and
Registers
MCLK
SYNC
AVDD1
AVDD2
AVDD3
AVDD4
DECAP
RBIAS
AGND
MCLK
VDRIVE
DVDD
DGND
FIR Filter
Engine
DRDY
CS
RESET
RD/WR
High performance 20-bit Sigma-Delta ADC
118dB SNR at 78kHz output data rate
100dB SNR at 2.5MHz output data rate
2.5 MHz maximum fully filtered output word rate
Programmable over-sampling rate (8x to 256x)
Flexible parallel 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
Filter bypass modes
Low power and power down modes
Synchronization of multiple devices via SYNC pin
DB0 - DB15
Figure 1.
APPLICATIONS
Data acquisition systems
Vibration analysis
Instrumentation
PRODUCT OVERVIEW
The AD7760 high performance 20-bit sigma delta analog to
digital converter combines wide input bandwidth and high
speed with the benefits of sigma delta conversion with
performance of 100dB SNR at 2.5MSPS making it ideal for high
speed data acquisition. Wide dynamic range combined with
significantly reduced anti-aliasing requirements simplify 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 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 to the
application. The AD7760 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 AD7760.
The reference voltage supplied to the AD7760 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 AD7760 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. PrN
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
© 2004 Analog Devices, Inc. All rights reserved.
AD7760
Preliminary Technical Data
TABLE OF CONTENTS
TABLE OF CONTENTS.................................................................. 2
Clocking the AD7760..................................................................... 14
AD7760—Specifications.................................................................. 3
Driving The AD7760...................................................................... 15
Timing Specifications....................................................................... 5
Using The AD7760..................................................................... 16
Timing Diagrams.............................................................................. 6
Bias Resistor Selection ............................................................... 16
Absolute Maximum Ratings............................................................ 7
Programmable FIR Filter............................................................... 17
ESD Caution.................................................................................. 7
Downloading a User-Defined Filter ............................................ 18
Pin Configuration and Functional Descriptions.......................... 8
Example Filter Download ......................................................... 18
Terminology .................................................................................... 10
AD7760 Registers ........................................................................... 20
Typical Performance Characteristics ........................................... 11
Non Bit-Mapped Registers........................................................ 21
Theory of Operation ...................................................................... 12
Outline Dimensions ....................................................................... 22
AD7760 Interface............................................................................ 13
Ordering Guide .......................................................................... 22
REVISION HISTORY
Rev. PrN | Page 2 of 22
Preliminary Technical Data
AD7760
AD7760—SPECIFICATIONS
Table 1. VDD1 = 2.5 V, VDD2 = 5 V, VREF = 4.096 V, TA = +25°C, Full Power Mode, unless otherwise noted
Parameter
DYNAMIC PERFORMANCE
Decimate by 256
Signal to Noise Ratio (SNR) 1
Spurious Free Dynamic Range (SFDR) 1
Total Harmonic Distortion (THD) 1
Intermodulation Distortion (IMD) 1
Decimate by 16
Signal to Noise Ratio (SNR) 1
Spurious Free Dynamic Range (SFDR) 1
Total Harmonic Distortion (THD) 1
Intermodulation Distortion (IMD) 1
Decimate by 8
Signal to Noise Ratio (SNR) 1
Spurious Free Dynamic Range (SFDR) 1
Total Harmonic Distortion (THD) 1
Intermodulation Distortion (IMD) 1
Intermodulation Distortion (IMD) 1
DC ACCURACY
Resolution
Integral Nonlinearity1
Differential Nonlinearity1
Offset Error1
Gain Error1
Offset Error Drift
Gain Error Drift
DIGITAL FILTER RESPONSE
Decimate by 8
Group Delay
Decimate by 16
Group Delay
Decimate by 128
Group Delay
ANALOG INPUT
Differential Input Voltage
DC Leakage Current
Input Capacitance
REFERENCE INPUT/OUTPUT
VREF Input Voltage
VREF Input DC Leakage Current
VREF Input Capacitance
POWER REQUIREMENTS
AVDD1 (Modulator Supply)
AVDD2 (General Supply)
AVDD3 (Diff-Amp Supply)
AVDD4 (Ref Buffer Supply)
DVDD
VDRIVE
Test Conditions/Comments
Specifcation
Unit
118
118
-100
-100
dB typ
dBFS typ
dB typ
dB typ
103
103
-100
-100
dB typ
dBFS typ
dB typ
dB typ
100
100
-100
-100
-100
dB typ
dBFS typ
dB typ
dB typ
dB typ
20
1
1
0.03
5
0.0006
0.1
Bits
LSB typ
LSB typ
% typ
LSB typ
% /°C
LSB /°C
MCLK = 40MHz
12
µS typ
MCLK = 40MHz
24
µS typ
MCLK = 24.576MHz
480
µS typ
Vin(+) – Vin(-), VREF = 2.5V
Vin(+) – Vin(-), VREF = 4.096V
±2
±3.25
±2
5
55
V pk-pk
V pk-pk
µA max
pF typ
pF typ
VDD3 = 3.3V
VDD3 = 5V
+2.5
+4.096
±1
5
Volts
Volts
µA max
pF max
±5%
±5%
+2.5
+5
+3.0/+5.5
+3.15/+5.25
+2.5
+1.65/+2.7
Volts
Volts
V min/max
V min/max
Volts
V min/max
MCLK = 24.576MHz, ODR = 48kHz, FIN = 1kHz Sine Wave
Non-harmonic
Input Amplitude = -6dB
MCLK = 40MHz, ODR = 1.25MHz, FIN =100kHz Sine Wave
Non-harmonic
Input Amplitude = -6dB
MCLK = 40MHz, ODR = 2.5MHz, FIN = 100kHz Sine Wave
Non-harmonic
Input Amplitude = -6dB
FIN = 100kHz Sine Wave
FIN = 1MHz Sine Wave
At 18 bits
Guaranteed monotonic to 20 bits
With internal buffer
With external buffer
±5%
Rev. PrN | Page 3 of 22
AD7760
Parameter
Full Power Mode
AIDD1 (Modulator)
AIDD2 (General)
AIDD4 (Reference Buffer)
Low Power Mode
AIDD1 (Modulator)
AIDD2 (General)
AIDD4 (Reference Buffer)
AIDD3 (Diff Amp)
DIDD
Standby Mode
AIDD1 (Modulator)
AIDD2 (General)
AIDD3 (Diff Amp)
AIDD4 (Reference Buffer)
DIDD
POWER DISSIPATION
Full Power Mode
Modulator (P1)
General (P2)
Reference Buffer (P4)
Low Power Mode
Modulator (P1)
General (P2)
Reference Buffer (P4)
Differential Amplifier (P3)
Digital Power
Standby Mode
1
Preliminary Technical Data
Test Conditions/Comments
Specifcation
Unit
AVDD4 = +5V
50
35
35
mA typ
mA typ
mA typ
AVDD4 = +5V
26
20
10
mA typ
mA typ
mA typ
AVDD3 = +5V, Both Modes
Both Modes
42
45
mA typ
mA typ
AVDD3 = +5V
AVDD4 = +5V
Clock Stopped
Clock Running
210
30
30
30
250
690
µA typ
nA typ
nA typ
nA typ
µA typ
µA typ
AVDD4 = +3.3V
AVDD4 = +5V
125
175
101
175
mW typ
mW typ
mW typ
mW typ
65
100
27
50
116
210
112.5
1.2
2.3
mW typ
mW typ
mW typ
mW typ
mW typ
mW typ
mW typ
mW typ
mW typ
AVDD4 = +3.3V
AVDD4 = +5V
AVDD3 = +3.3V
AVDD3 = +5V
Clock Stopped
Clock Running
See Terminology
Rev. PrN | Page 4 of 22
Preliminary Technical Data
AD7760
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
fMCLK
Unit
MHz min
MHz max
MHz min
MHz max
typ
Description
Applied Master Clock Frequency
t11
Limit at TMIN, TMAX
12.288
80
12.288
20
0.5 × tICLK
t2
10
nS min
DRDY Falling Edge to CS falling Edge
t3
2
nS min
RD/WR Setup Time to CS Falling Edge
t4
t5
10
tICLK
nS typ
min
Data Access Time
CS Low Pulse Width
t6
tICLK
min
CS High Pulse Width Between Reads
t7
2
nS min
RD/WR Hold Time to CS Rising Edge
t8
t9
10
0.5 × tICLK
nS max
typ
Bus Relinquish Time
DRDY High Period
t10
0.5 × tICLK
typ
DRDY Low Period
t11
t12
15
TBD
nS typ
xS min
Data Access Time
Data Valid Prior to DRDY Rising Edge
t13
TBD
xS min
Data Valid After DRDY Rising Edge
t14
t15
10
tICLK
nS max
xS min
Bus Relinquish Time
CS Low Pulse Width
t16
tICLK
xS min
CS High Period Between Address and Data
t17
t18
10
10
nS min
nS min
Data Setup Time
Data Hold Time
fICLK
1
Internal Modulator Clock Derived from MCLK.
DRDY Pulse Width
tICLK = 1/fICLK
Rev. PrN | Page 5 of 22
AD7760
Preliminary Technical Data
TIMING DIAGRAMS
t5
t1
t6
t2
t7
t3
t4
t8
Figure 2. Parallel Interface Timing Diagram
t9
t11
t12
t10
t13
t14
Figure 3. 20MHz Modulator Data Output Mode
t15
t17
t16
t18
Figure 4. AD7760 Register Write
Rev. PrN | Page 6 of 22
Preliminary Technical Data
AD7760
ABSOLUTE MAXIMUM RATINGS
Table 3. TA = 25°C, unless otherwise noted.
Parameters
VDD to GND
VIN+ to GND
VIN– to GND
Digital input voltage to GND
Digital output voltage to GND
VREF to GND
Input current to any pin except supplies1
Operating temperature range
Commercial (A, B version)
Storage temperature range
Junction temperature
TQFP Exposed Paddle Package
θJA thermal impedance
θJC thermal impedance
CSP Package
θJA thermal impedance
θJC thermal impedance
Lead temperature, soldering
Vapor phase (60 secs)
Infrared (15 secs)
ESD
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.
Rating
TBD
TBD
TBD
TBD
TBD
TBD
TBD
−40°C to +85°C
−65°C to +150°C
150°C
92.7 °C/W
5.1 °C/W
26.7 °C/W
30 °C/W
215°C
220°C
TBD kV
1
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.
Rev. PrN | Page 7 of 22
AD7760
Preliminary Technical Data
64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49
12
37 RESET
13
36 SYNC
14
35 DGND
34 AGND
33 AV
DD1
37 DB11
38 DB10
40 DB8
41 DB7
42 DB6
43 DB5
39 DB9
30 CS
29 RD/WR
28 DRDY
27 RESET
26 SYNC
25 AVDD1
AGND 24
11
38 DRDY
31 DVDD
AVDD2 10
AVDD2 11
RBIAS 12
DECAP2 21
39 RD/WR
AGND 16
44 DB4
TOP VIEW
(Not to Scale)
40 CS
15
45 DB3
AD7760
REFGND 7
VREF+ 8
AVDD4 9
41 DV
DD
TOP VIEW
(Not to Scale)
46 DB2
DECAP1 6
43 DGND
VIN- 19
AVDD2 20
DECAP1 8
REFGND 9
VREF+ 10
AGND
AVDD4
AGND
AVDD2
AVDD2
44 VDRIVE
42 DGND
AD7760
33 DB15
32 VDRIVE
AGND 23
45 DB15
DECAP3 22
46 DB14
AGND 5
AVDD1 6
AGND 7
35 DB13
34 DB14
MCLK 3
AVDD2 4
AVDD1 5
47 DB13
VOUTA1+ 16
AVDD3 17
VIN+ 18
PIN 1
IDENTIFIER
36 DB12
PIN 1
IDENTIFIER
VINA1+ 13
VINA1- 14
VOUTA1- 15
MCLK 2
MCLK 3
AVDD2 4
VDRIVE 1
MCLK 2
48 DB12
DGND 1
47 DB1
48 DB0
DB11
DB10
DB9
DGND
DB8
DB7
DB6
DB5
DB4
DB3
DB2
DB1
DB0
DGND
DGND
VDRIVE
PIN CONFIGURATION AND FUNCTIONAL DESCRIPTIONS
AGND
AGND
DECAP3
DECAP2
AGND
AVDD2
VIN+
VIN-
AVDD3
AGND
VOUTA1VOUTA1+
VINA1+
VINA1-
AGND
RBIAS
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Figure 6. 48-PIN LFCSP Pin Configuration
Figure 5. 64-Lead TQFP Pin Configuration
Table 4. Pin Function Descriptions
TQFP Pin
Number
6, 33
CSP Pin
Number
5, 25
Pin Mnemonic
Description
AVDD1
4, 14, 15,
27
24
4, 10, 11,
20
17
AVDD2
12
9
AVDD4
5, 7, 11,
13, 16, 18,
23, 28, 31,
32, 34
9
41
23, 24,
Paddle
AGND
+2.5V power supply for modulator. These pins should be decoupled to AGND with 100nF and
10µF capacitors on each pin.
+5V power supply. These pins should be decoupled to AGND with TBD nF and TBD µF
capacitors on each pin.
+3.3V to +5V power supply for differential amplifier. These pins should be decoupled to AGND
with a 100nF capacitor.
+3.3V to +5V power supply for reference buffer. This pin should be decoupled to AGND with a
10nF capacitor in series with a 22Ω resistor.
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 AD7760.
7
31
REFGND
DVDD
44, 63
1, 32
VDRIVE
1, 35, 42,
43, 53, 62,
64
Paddle
DGND
AVDD3
Reference Ground. Ground connection for the reference voltage.
+2.5V power supply for digital circuitry and FIR filter. This pin should be decoupled to DGND
with a 470nF capacitor.
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 470nF capacitor.
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 AD7760.
Rev. PrN | Page 8 of 22
Preliminary Technical Data
AD7760
TQFP Pin
Number
19
20
21
22
25
26
10
CSP Pin
Number
13
14
15
16
18
19
8
Pin Mnemonic
Description
VINA1+
VINA1VOUTA1VOUTA1+
VIN+
VINVREF+
8
29
30
17
6
21
22
12
DECAP1
DECAP2
DECAP3
RBIAS
45-52,
54-61
33-48
DB15 – DB0
37
27
RESET
3
3
MCLK
2
36
2
26
MCLK
SYNC
39
29
RD/WR
38
28
DRDY
40
30
CS
Positive Input to Full-Power Differential Amplifier 1.
Negative Input to Full-Power Differential Amplifier 1.
Negative Output from Full-Power Differential Amplifier 1.
Positive Output from Full-Power Differential Amplifier 1.
Positive Input to the Modulator.
Negative Input to the Modulator.
Reference Input. The input range of this pin is determined by the reference buffer supply
voltage (AVDD4). See Reference Section for more details.
Decoupling Pin. A 100nF capacitor must be inserted between this pin and AGND.
Decoupling Pin. A TBD µF capacitor must be inserted between this pin and AGND.
Decoupling Pin. A TBD µF capacitor must be inserted between this pin and AGND.
Bias Current setting pin. A resistor must be inserted between this pin and AGND. For more
details on this, see the Bias Resistor Section.
16-bit bi-directional data bus. These are three-state pins that are controlled by the CS and RD
/WR pins. The operating voltage for these pins is determined by the VDRIVE voltage. See
Interfacing Section for more details.
A falling edge on this pin resets all internal digital circuitry. Holding this pin lows keeps the
AD7760 in a reset state.
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.
Master Clock ground sensing pin.
Synchronization Input. A falling edge on this pin resets the internal filter. This can be used to
synchronize multiple devices in a system.
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 will take place. If this pin
is high and CS is low, a write will occur. See AD7760 Interface Section for more details.
Data Ready Output. Each time that new conversion data is available, an active low pulse,
½ ICLK period wide, is produced on this pin. See AD7760 Interface Section for further details.
Chip Select Input. Used in conjunction with the RD/WR pin to read and write data to and from
the AD7760. See AD7760 Interface Section for further details.
Rev. PrN | Page 9 of 22
AD7760
Preliminary Technical Data
TERMINOLOGY
Signal to (Noise + Distortion) Ratio
The measured ratio of signal to (noise + distortion) at the
output of the ADC. The signal is the rms amplitude of the
fundamental. Noise is the sum of all nonfundamental signals up
to half the sampling frequency (fS/2), excluding dc. The ratio is
dependent on the number of quantization levels in the
digitization process; the more levels, the smaller the
quantization noise. The theoretical signal to (noise + distortion)
ratio for an ideal N-bit converter with a sine wave input is given
by
Signal to (Noise + Distortion) = (6.02 N + 1.76 ) dB
Thus, for an 18-bit converter, this is 110.12dBs and for a 20-bit
converter, 122.16 dB.
Total Harmonic Distortion (THD)
The ratio of the rms sum of harmonics to the fundamental. For
the AD7760, it is defined as
THD (dB ) = 20 log
V22 + V32 + V42 + V52 + V62
V1
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 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, 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.
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.
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
Offset Error
The deviation of the first code transition (000...000 to 000...001)
from the ideal (that is, AGND + 1 LSB).
Gain Error
The deviation of the last code transition (111...110 to 111...111)
from the ideal (that is, VREF − 1 LSB), after the offset error has
been adjusted out.
Power Supply Rejection Ratio (PSRR)
The ratio of the power in the ADC output at full-scale
frequency, f, to the power of a 100 mV p-p sine wave applied to
the ADC VDD supply of frequency fs. The frequency of this
input varies from 1 kHz to 1 MHz.
PSRR (dB ) = 10 log (Pf Pfs )
Pf is the power at frequency f in the ADC output; Pfs is the
power at frequency fs in the ADC output.
Rev. PrN | Page 10 of 22
Preliminary Technical Data
AD7760
TYPICAL PERFORMANCE CHARACTERISTICS
000
000
000
000
000
ALL CAPS ( Init ial cap)
ALL CAPS ( Init ial cap)
Default Conditions: TA = 25°C, TBD, unless otherwise noted.
TBD
000
000
000
000
000
TBD
000
000
000
000
000
ALL CAPS ( Init ial cap)
000
000
000
000
000
Figure 7. TBD
000
ALL CAPS ( Init ial cap)
ALL CAPS ( Init ial cap)
000
000
000
000
000
TBD
000
000
TBD
000
000
000
000
000
000
ALL CAPS ( Init ial cap)
000
000
000
000
000
Figure 8. TBD
000
000
000
000
TBD
000
000
000
ALL CAPS ( Init ial cap)
Figure 11. TBD
000
ALL CAPS ( Init ial cap)
ALL CAPS ( Init ial cap)
000
Figure 10. TBD
000
000
000
000
000
000
000
000
000
000
000
ALL CAPS ( Init ial cap)
000
TBD
000
000
000
000
000
ALL CAPS ( Init ial cap)
000
000
Figure 9. TBD
000
000
000
000
000
ALL CAPS ( Init ial cap)
Figure 12. TBD
Rev. PrN | Page 11 of 22
AD7760
Preliminary Technical Data
THEORY OF OPERATION
The AD7760 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 13a). 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 13b).
The digital filtering which follows the modulator removes the
large out-of-band quantization noise (Figure 13c) 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 AD7760 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 AD7760 at a large range of data rates. Multi-bit data
from the modulator can be obtained at a rate of 20MHz. The
first filter receives data from the modulator at 20MHz where it
is decimated by four to output data at 5MHz. This partially
filtered data can also be output at this stage. The second filter
allows the decimation rate to be chosen from 2x to 32x or to be
QUANTIZATION NOISE
fICLK/2
BAND OF INTEREST
a.
NOISE SHAPING
fICLK/2
BAND OF INTEREST
b.
DIGITAL FILTER CUTOFF FREQUENCY
fICLK/2
BAND OF INTEREST
c.
Figure 13. Sigma-Delta ADC
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 X 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
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
20 MHz
12.288MHz
12.288MHz
12.288MHz
12.288MHz
Filter 1
Filter 2
Filter 3
Data State
Bypassed
4x
4x
4x
4x
4x
4x
4x
4x
4x
4x
4x
4x
4x
4x
4x
4x
Bypassed
Bypassed
Bypassed
2x
2x
4x
4x
8x
8x
16x
16x
32x
32x
8x
16x
32x
32x
Bypassed
Bypassed
2x
Bypassed
2x
Bypassed
2x
Bypassed
2x
Bypassed
2x
Bypassed
2x
2x
2x
Bypassed
2x
Unfiltered
Partially Filtered
Fully Filtered
Partially Filtered
Fully Filtered
Partially Filtered
Fully Filtered
Partially Filtered
Fully Filtered
Partially Filtered
Fully Filtered
Partially Filtered
Fully Filtered
Fully Filtered
Fully Filtered
Partially Filtered
Fully Filtered
Computation
Delay
0
0.325
1.075
1.35
1.625
1.725
1.775
2.6
2.25
4.175
3.1
7.325
4.65
3.66
5.05
11.92
7.57
Rev. PrN | Page 12 of 22
Filter Delay
0
1.2µS
10.8µS
3.6µS
22.8µS
6µS
44.4µS
10.8µS
87.6µS
20.4µS
174µS
39.6µS
346.8µS
142.6µS
283.2µS
64.45µS
564.5µS
Passband
Bandwidth
(10 MHz)
1.35 MHz
1 MHz
562.5 kHz
500 kHz
281.25 kHz
250 kHz
140.625 kHz
125 kHz
70.3125 kHz
62.5 kHz
35.156 kHz
31.25 kHz
76.8 kHz
38.4 kHz
21.6 kHz
19.2 kHz
Output Data
Rate (ODR)
20 MHz
5 MHz
2.5 MHz
2.5 MHz
1.25 MHz
1.25 MHz
625 kHz
625 kHz
312.5 kHz
312.5 kHz
156.25 kHz
156.25 kHz
78.125 kHz
192 kHz
96 kHz
96 kHz
48 kHz
Preliminary Technical Data
AD7760
AD7760 INTERFACE
Reading Data
The AD7760 uses a 16-bit bi-directional parallel interface. This
interface is controlled by the RD/WR and CS pins. There are
two read operating modes depending on the output data rate.
When the AD7760 is outputting data at 5MSPS or less, the
interface operates in a conventional mode as shown in Figure 2.
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 these lines go low, the databus 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
period of TBD ns before the second read is performed. This
second read will contain the 8 Least Significant Bits (LSBs) of
the conversion result along with 7 status bits. These status bits
are shown in Table 6. The Cal bit is set to a 1 if a calibration has
been performed. Table 14 contains descriptions of the other
status bits.
Table 6. Status Bits During Data Read
D7
DValid
D0
Ovr
UFilt
LPwr
FiltOk
DLOk
Cal
0
Shortly after RD/WR and CS return high, the databus will
return to a high impedance state. Both read operations must be
completed before a new conversion result is available as the new
result will overwrite the contents on the output register. If a
DRDY pulse occurs during a read operation, the data read will
be invalid.
When the AD7760 is operating in modulator data output mode,
i.e. Output Data Rate at 20MHz, a different interfacing scheme
is necessary. To obtain data from the AD7760 in this mode, both
RD/WR and CS lines must be held low. This will bring the
databus out of its high impedance state. Figure 3 shows the
20MHz Output Data Rate operation. A DRDY pulse is
generated for each word and the data is valid on the rising edge
of the DRDY pulse. This DRDY pulse could 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 will
stop outputting data and the databus will return to high
impedance.
Sharing The Parallel Bus
By its nature, the high accuracy of the AD7760 make 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 are isolated from the system databus by means of a
latch or buffer to ensure that there is no digital activity on the
D0-D15 pins that is not 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.
Writing To The AD7760
After a reset, only a single write operation to power up the
AD7760 is necessary to start the part converting on default
settings. While the AD7760 is configured to convert analog
signals with the default settings on reset, there are many features
and parameters on this part that the user can change by writing
to the device. As some of the programmable registers are 16 bits
wide, to program a register requires two write operations. The
first write contains the register address while the second write
contains the register data. There is an exception to this when a
user filter is being downloaded to the AD7760. This is dealt
with in detail in the following section. The AD7760 Registers
section contains the register addresses and further details.
Figure 4 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 TBD ns. The register address is latched during this period.
The CS line is brought high again for a minimum of TBD ns
before the register data is put onto the databus. 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 again. This also provides a
method to get back to a known situation if the user somehow
loses track whether the next write is an address or data.
It is envisaged that the AD7760 will be 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 pass before valid data will be output from the
AD7760.
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 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 AD7760
Registers section.
Rev. PrN | Page 13 of 22
AD7760
Preliminary Technical Data
CLOCKING THE AD7760
The AD7760 requires an external low jitter clock source. This
signal is applied to the MCLK and MCLK pins. An internal
clock signal (ICLK) is derived from the MCLK input signal.
This ICLK controls all the internal operation of the AD7760.
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[1:0] = 10)
2.
ICLK = MCLK / 2 (CDIV[1:0] = 00)
3.
ICLK = MCLK / 4 (CDIV[1:0] = 01)
t j ( RMS ) =
256
= 133 ps
2 × π × 19.2 × 103 × 106
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 14 and Figure 15 illustrate this point showing the
maximum slew rate of a sine wave of the same frequency but
with different amplitudes.
These options are selected from the control register (See
Register Section for further details). On power-up, the default is
ICLK = MCLK / 4 to ensure that the part can handle the
maximum MCLK frequency of 80MHz. 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, 96 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:
Figure 14. Maximum Slew Rate of Sine Wave with Amplitude of 2V Pk-Pk
OSR
t j ( RMS ) =
2 × π × f IN × 10
SNR ( dB )
20
OSR = Over-sampling ratio = f ICLK
ODR
fIN = Maximum Input Frequency
SNR(dB) = Target SNR.
Taking an example from Table 5:
ODR = 2.5MHz, fICLK = 20MHz, fIN (max) = 1MHz, SNR =
108dB
t j ( RMS ) =
8
= 1.79 ps
2 × π × 106 ×105.4
Figure 15. Maximum Slew Rate of Same Frequency Sine Wave with
Amplitude of 1V Pk-Pk
This is the maximum allowable clock jitter for a full-scale
1MHz 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
Rev. PrN | Page 14 of 22
Preliminary Technical Data
AD7760
DRIVING THE AD7760
The AD7760 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 antialias filter as shown in Figure 16. Any additional filtering should
be carried out in previous stages using low noise, highperformance op-amps such as the AD8021.
+2.5V
+3.685V
0V
+2.048V
A
0.410V
-2.5V
+2.5V
VIN+
+3.685V
B
0V
+2.048V
-2.5V
0.410V
VIN-
Figure 17. Differential Amplifier Signal Conditioning
Suitable component values for the first order filter are listed in
Table 7. Using the first row as an example would yield a 10dB
attenuation at the first alias point of 19MHz.
CFB
RFB
RIN
A
VINA1
CS
B
VIN+
RIN
To obtain maximum performance from the AD7760, it is
advisable to drive the ADC with differential signals. However, it
is possible to drive the AD7760 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 18 shows how a bipolar
single-ended signal biased around ground can be used to drive
the AD7760 with the use of an external op-amp such as the
AD8021.
CFB
RFB
2R
VIN
CFB
Figure 16. Differential Amplifier Configuration
RFB
RFB
AD8021
R
Table 7. Full Power Component Values
RIN
2R
RIN
CS
VIN+
RIN
ODR
VREF
CS
2.5MHz
4.096v
1kΩ
655 Ω
5.6pF
33pF
2.5MHz
2.5v
TBD Ω
TBD Ω
TBD
TBD pF
48kHz
4.096v
TBD Ω
TBD Ω
TBD
TBD pF
48kHz
2.5v
TBD Ω
TBD Ω
TBD
TBD pF
VINA1
RFB
CFB
Figure 17 shows the signal conditioning that occurs using the
circuit in Figure 16 with a ±2.5v input signal biased around
ground using the component values and conditions in the first
row of Table 7. 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.
Rev. PrN | Page 15 of 22
CFB
Figure 18. Single Ended to Differential Conversion
AD7760
Preliminary Technical Data
USING THE AD7760
BIAS RESISTOR SELECTION
The following is the recommended sequence for powering up
and using the AD7760.
The AD7760 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 100kΩ and for a 4.096V reference, 160kΩ.
1.
Apply Power
2.
Start clock oscillator, applying MCLK
3.
Take 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 amplifiers as required. The correct
Clock Divider (CDIV[1:0]) ratio should be
programmed here also.
6.
Write to Control Register 1 to set up the Output Data
Rate.
7.
Take SYNC low for a minimum of 2 MCLK 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
group delay of the filter has passed. When this has
occurred, the DVALID bit read with the data LSW 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. An internal calibration sequence can also
be initiated at this point.
Rev. PrN | Page 16 of 22
Preliminary Technical Data
AD7760
PROGRAMMABLE FIR FILTER
As previously mentioned, the third FIR filter on the AD7760 is
user programmable. The default coefficients that are loaded on
reset are given in Table 8. This gives a frequency response
shown in Figure 19. The frequencies quoted in Figure 19 scale
directly with the Output Data Rate.
•
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.
Table 8. Default Filter Coefficients
#
Dec. Value
Hex
Value
#
0
53656736
332BCA0
24
Dec.
Value
700847
1
25142688
17FA5A0
25
-70922
2
-4497814
444A196
26
-583959
408E917
3
-11935847
4B62067
27
-175934
402AF3E
4
-1313841
4140C31
28
388667
5EE3B
Hex Value
AB1AF
401150A
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
0
Passband Ripple = 0.05 dB
-0.1dB Frequency = 1.004MHz
-3dB Frequency = 1.06MHz
Stopband = 1.25MHz
-20
-40
Amplitude (dB)
5
-60
-80
-100
-120
-140
-160
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, in 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.
0
500
1000
1500
2000
2500
Frequency (kHz)
Figure 19. Default Filter Frequency Response (2.5MHz ODR)
The procedure for downloading a user-defined filter is detailed
in the Downloading a User-Defined Filter section.
If a user does wish to create their own filter then the following
should be noted:
Rev. PrN | Page 17 of 22
AD7760
Preliminary Technical Data
As previously mentioned, the filter coefficients are 27 bits in
length; one sign and 26 magnitude bits. Since the AD7760 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 AD7760 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.
2.
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
9).
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.
b.
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 20.
10
0
-10
Amplitude (dB)
DOWNLOADING A USER-DEFINED FILTER
-20
-30
-40
-50
-60
-70
-80
0
300
400
500
600
Figure 20. 24-Tap FIR Frequency Response
The coefficients for the filter are listed in Table 10. The
coefficients are in 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 10. 24-Tap FIR Coefficients
Read the Status Register checking the
DL_OK bit.
Coeff
Start reading data and observe the status of
the DL_OK bit.
Num Coeffs
Default
6
12
18
24
30
36
42
48
200
Frequency (kHz)
Table 9. Filter Length Values
FLEN[3:0]
0000
0001
0011
0101
0111
1001
1011
1101
1111
100
Filter Length
Default
12
24
36
48
60
72
84
96
Rev. PrN | Page 18 of 22
1
Raw
0.365481974
Scaled
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
AD7760
Table 11 shows the Hex values (in sign and magnitude format)
that are downloaded to the AD7760 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 11. Filter Hex Values
Coeff
0x0001 Address of Control Register 1
0x8079 Control Reg Data; DL Filter, Set Filter Length = 24, Set
Output Data Rate = 1.25MHz
0x032B First Coefficient, Word 1
0x9688 First Coefficient, Word 2
Word 2
Word 1
0x01BF Second Coefficient, Word 1
Byte 1
Byte 2
Byte 3
1
03
2B
96
88
2
01
BF
18
3C
3
00
15
66
26
4
04
A8
1E
6A
0x0404 Twelfth (Final) Coefficient, Word 1
5
04
5F
2A
96
0x3B5F Final Coefficient, Word 2
6
00
2C
49
C2
7
00
50
E9
F6
8
00
10
DB
D8
9
04
2F
DE
54
Wait TBD xS for AD7760 to fill remaining unused coefficients
with zeros.
10
04
37
44
5A
0x0001 Address of Control Register
11
04
1B
7D
6E
12
04
04
3B
5F
Byte 4
What follows is a list of 16-bit words that the user would write
to the AD7760 to set up the ADC and download this filter
assuming an output data rate of 1.25MHz has already been
selected.
0x183C Second Coefficient, Word 2
…
…
0x0E6B Checksum
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.
Rev. PrN | Page 19 of 22
AD7760
Preliminary Technical Data
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 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 12. Control Register 1 (Address 0x0001, Default Value 0x001A)
MSB
DL
Filt
LSB
RD
Ovr
RD
Gain
Bit
15
Mnemonic
DL Filt1
14
RD Ovr1,2
13
12
11
10
9
RD Gain1,2
RD Off1,2
RD Stat1,2
CAL1
SYNC1
8-5
4
3
FLEN3:0
BYP F3
BYP F1
2-0
DEC2:0
1
2
RD
Off
RD
Stat
CAL
SYNC
FLEN3
FLEN2
FLEN1
FLEN0
BYP F3
BYP F1
DEC2
DEC1
DEC0
Comment
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.
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.
Read Gain. If this bit has been set, the next read operation will output the contents of the digital Gain Register.
Read Offset. If this bit has been set, the next read operation will output the contents of the digital Offset Register.
Read Status. If this bit has been set, the next read operation will output the contents of the Status Register.
Calibration. Setting this bit will initiate an internal calibration routine. This routine will take 14mS with a 20MHz ICLK.
Synchronize. Setting this bit will initiate in internal synchronisation routine. Setting this bit simultaneously on multiple
devices will synchronize all filters.
Filter Length Bits. These bits must be set when the DL Filt bit is set and before a user defined filter is downloaded.
Bypass Filter 3. If this bit is a 0, Filter 3 (Programmable FIR) will be bypassed.
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.
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.
Bits 15-9 are all self clearing bits.
Only one of the bits 14-11 may be set in any write operation as they all determine the contents of the next read operation
Table 13. Control Register 2 (Address 0x0002, Default Value 0x009B)
MSB
0
LSB
0
0
Bit
5-4
Mnemonic
CDIV1:0
3
2
PD
LPWR
1
0
D1PD
0
0
0
0
0
0
0
CDIV1
CDIV0
PD
LPWR
1
D1PD
Comment
Clock Divider Bits. These set the divide ratio of the MCLK signal to produce the internal ICLK. Setting CDIV[1:0] = 00
divides the MCLK by 2, setting CDIV[1:0] = 01 divides MCLK by 4. If CDIV[1:0] = 10 then the MCLK frequency is equal to
the ICLK. CDIV[1:0] = 11 is not allowed.
Power Down. Setting this bit powers down the AD7760 reducing the power consumption to TBD µW.
Low Power. If this bit is set, the AD7760 is operating in a low power mode. The power consumption is reduced for a 6dB
reduction in noise performance.
Write a ‘1’ to this bit.
Differential Amplifier Power Down. Setting this bit powers down the on-chip differential amplifier.
Rev. PrN | Page 20 of 22
Preliminary Technical Data
AD7760
Table 14. Status Register (Read Only)
MSB
LSB
PART
1
PART
0
DIE
2
Bit
15,14
13-11
10
9
8
7
Mnemonic
PART1:0
DIE2:0
DVALID
LPWR
OVR
DL OK
6
Filter OK
5
4
3
2-0
U Filter
BYP F3
BYP F1
DEC2:0
DIE
1
DIE
0
DVALID
LPWR
OVR
DL
OK
Filter
OK
U
Filter
BYP F3
BYP F1
DEC2
DEC1
DEC0
Comment
Part Number. These bits will be constant for the AD7760.
Die Number. These bits will reflect the current AD7760 die number for identification purposes within a system.
Data Valid. This bit corresponds to the DVALID bit in the status word output in the second 16-bit read operation.
Low Power. If the AD7760 is operating in Low Power Mode, this bit is set to a 1.
If the current analog input exceeds the current overrange threshold, this bit will be set.
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.
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.
If a user-defined filter is in use, this bit is set.
Bypass Filter 3. If Filter 3 is bypassed by setting the relevant bit in Control Register 1, this bit is also set.
Bypass Filter 1. If Filter 1 is bypassed by setting the relevant bit in Control Register 1, this bit is also set.
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.
Rev. PrN | Page 21 of 22
AD7760
Preliminary Technical Data
OUTLINE DIMENSIONS
0.60 (0.024)
0.42 (0.017)
0.24 (0.009)
0.60 (0.024)
0.42 (0.017)
37
36
0.24 (0.009)
7.0 (0.276) BSC SQ
PIN 1
INDICATOR
0.25 (0.010)
MIN
48
1
6.75 (0.266)
BSC SQ
TOP
VIEW
5.25 (0.207)
5.10 (0.201) SQ
4.95 (0.195)
BOTTOM
VIEW
25
24
0.50 (0.020)
0.40 (0.016)
0.30 (0.012)
12
13
5.5 (0.217)
REF
0.70 (0.028) MAX
12°MAX
0.65 (0.026) NOM
0.90 (0.035) MAX
0.05 (0.002)
0.85 (0.033) NOM
0.01 (0.0004)
0.0 (0.0)
0.30 (0.012) 0.50 (0.020) 0.20 (0.008)
REF
0.23 (0.009)
BSC
0.18 (0.007)
Figure 21. 48-Lead Frame Chip Scale Package [LFCSP] (CP-48)—Dimensions shown in millimeters
1.60 (0.063) MAX
0.15(0.006)
0.05(0.002)
0.60 ± 0.15
(0.024 ± 0.006)
12o
TYP
12.0(0.47) BSC
10.0(0.39) BSC
64
49
48
1
SEATING
PLANE
6.0(0.235)
BSC
TOP VIEW
33
32
16
0o
3.5o ± 3.5o
17
0.50 (0.02)
BSC
0.22 ± 0.05
(0.0087 ± 0.002)
Figure 22. 64-Lead Thin Quad Flat Pack (Exposed Paddle) [TQFP] (SV-64)—Dimensions shown in millimeters
ORDERING GUIDE
Model
AD7760BCP
AD7760BSV
Temperature Range
–40°C to +85°C
–40°C to +85°C
Package Description
Lead Frame Chip Scale Package
Thin Quad Flat Pack, Exposed Paddle
© 2004 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
Printed in the U.S.A.
PR04975-0-6/04(PrN)
Rev. PrN | Page 22 of 22
Package Option
CP-48
SV-64