AD73322LASTZ [ADI]
IC SPECIALTY TELECOM CIRCUIT, PQFP44, PLASTIC, MS-026-BEA, LQFP-44, Telecom IC:Other;
| 型号： | AD73322LASTZ |
| 厂家： | 
ADI |
| 描述： | IC SPECIALTY TELECOM CIRCUIT, PQFP44, PLASTIC, MS-026-BEA, LQFP-44, Telecom IC:Other 电信 电信集成电路 |
| 文件： | 总48页 (文件大小：729K) |
| 中文： | 中文翻译 | 下载： | 下载PDF数据表文档文件 |
Low Cost, Low Power CMOS
General-Purpose Dual Analog Front End
AD73322L
FUNCTIONAL BLOCK DIAGRAM
FEATURES
Two 16-bit A/D converters
Two 16-bit D/A converters
Programmable input/output sample rates
78 dB ADC SNR
AVDD1 AVDD2 DVDD
AD73322L
VFBP1
SDI
SDIFS
VINP1
VINN1
VFBN1
ADC CHANNEL 1
78 dB DAC SNR
64 kHz maximum sample rate
−90 dB crosstalk
Low group delay (25 µs typ per ADC channel, 50 µs typ per
DAC channel)
VOUTP1
VOUTN1
DAC CHANNEL 1
REFERENCE
SCLK
SE
RESET
REFOUT
REFCAP
SPORT
Programmable input/output gain
Flexible serial port allows up to 4 dual codecs to be
connected in cascade, giving 8 I/O channels
VFBP2
VINP2
VINN2
VFBN2
ADC CHANNEL 2
MCLK
Single-supply operation (2.7 V to 3.3 V)
50 mW typ power consumption at 3.0 V
Temperature range: −40°C to +105°C
On-chip reference
SDOFS
SDO
VOUTP2
VOUTN2
DAC CHANNEL 2
AGND1 AGND2 DGND
28-lead SOIC, TSSOP, and 44-lead LQFP packages
Figure 1.
APPLICATIONS
General-purpose analog I/O
Speech processing
Cordless and personal communications
Telephony
Active control of sound and vibration
Data communications
Wireless local loop
GENERAL DESCRIPTION
The A/D and D/A conversion channels feature programmable
input/output gains with ranges of 38 dB and 21 dB, respectively.
An on-chip reference voltage allows single-supply operation.
The AD73322L is a dual front-end processor for general-
purpose applications, including speech and telephony. It
features two 16-bit A/D conversion channels and two 16-bit
D/A conversion channels. Each channel provides 78 dB signal-
to-noise ratio over a voice-band signal bandwidth. It also
features an input-to-output gain network in both the analog
and digital domains. This is featured on both codecs and can
be used for impedance matching or scaling when interfacing to
subscriber line interface circuits (SLICs).
The sampling rate of the codecs is programmable with four
separate settings offering 64 kHz, 32 kHz, 16 kHz, and 8 kHz
sampling rates (from a master clock of 16.384 MHz).
A serial port (SPORT) allows easy interfacing of single or
cascaded devices to industry-standard DSP engines. The
SPORT transfer rate is programmable to allow interfacing to
both fast and slow DSP engines.
The AD73322L is particularly suitable for a variety of appli-
cations in the speech and telephony area, including low bit rate,
high quality compression, speech enhancement, recognition,
and synthesis. The low group delay characteristic of the part
makes it suitable for single or multichannel active control
applications.
The AD73322L is available in 28-lead SOIC, 28-lead TSSOP,
and 44-lead LQFP packages.
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
Fax: 781.326.8703
www.analog.com
© 2004 Analog Devices, Inc. All rights reserved.
AD73322L
TABLE OF CONTENTS
Specifications..................................................................................... 4
Sample Rate Divider................................................................... 19
DAC Advance Register.............................................................. 20
Control Register A ..................................................................... 21
Control Register B...................................................................... 21
Control Register C...................................................................... 21
Control Register D ..................................................................... 22
Control Register E...................................................................... 22
Control Register F ...................................................................... 22
Control Register G ..................................................................... 23
Control Register H ..................................................................... 23
Operation......................................................................................... 24
Resetting the AD73322L ........................................................... 24
Power Management ................................................................... 24
Operating Modes........................................................................ 24
Program (Control) Mode.......................................................... 24
Data Mode................................................................................... 25
Mixed Program/Data Mode...................................................... 25
Digital Loop-Back Mode........................................................... 25
SPORT Loop-Back Mode.......................................................... 25
Analog Loop-Back Mode .......................................................... 26
Interfacing ....................................................................................... 27
Cascade Operation..................................................................... 27
Performance .................................................................................... 29
Encoder Section.......................................................................... 29
Encoder Group Delay ................................................................ 30
Decoder Section ......................................................................... 30
On-Chip Filtering....................................................................... 31
Decoder Group Delay................................................................ 31
Design Considerations................................................................... 32
Analog Inputs ............................................................................. 32
Interfacing to an Electret Microphone .................................... 34
Current Summary......................................................................... 6
Signal Ranges ................................................................................ 7
Timing Characteristics ................................................................ 7
Timing Diagrams.......................................................................... 8
Absolute Maximum Ratings............................................................ 9
ESD Caution.................................................................................. 9
Pin Configurations and Function Descriptions ......................... 10
Terminology .................................................................................... 12
Abbreviations .............................................................................. 12
Typical Performance Characteristics and Functional Block
Diagram ........................................................................................... 13
Functional Descriptions ................................................................ 14
Encoder Channels ...................................................................... 14
Programmable Gain Amplifier................................................. 14
ADC ............................................................................................. 14
Analog Sigma-Delta Modulator ............................................... 14
Decimation Filter........................................................................ 15
ADC Coding ............................................................................... 15
Decoder Channel........................................................................ 16
DAC Coding................................................................................ 16
Interpolation Filter ..................................................................... 16
Analog Smoothing Filter and PGA.......................................... 16
Differential Output Amplifiers................................................. 16
Voltage Reference ....................................................................... 16
Analog and Digital Gain Taps................................................... 17
Digital Gain Tap.......................................................................... 18
Serial Port (SPORT) ................................................................... 18
SPORT Overview........................................................................ 18
SPORT Register Maps................................................................ 19
Master Clock Divider................................................................. 19
Serial Clock Rate Divider .......................................................... 19
Rev. A | Page 2 of 48
AD73322L
Analog Output.............................................................................34
Differential-to-Single-Ended Output.......................................35
Digital Interfacing.......................................................................35
Cascade Operation......................................................................35
Grounding and Layout ...............................................................36
DSP Programming Considerations ..............................................37
DSP SPORT Configuration .......................................................37
DSP SPORT Interrupts...............................................................37
Mixed-Mode Operation.............................................................37
Interrupts .....................................................................................37
Initialization.................................................................................38
Running the AD73322L with ADCs or DACs in Power-Down
.......................................................................................................38
DAC Timing Control Example .....................................................40
Configuring an AD73322L to Operate in Data Mode ...............41
Configuring an AD73322L to Operate in Mixed Mode ............43
Outline Dimensions........................................................................46
Ordering Guide ...........................................................................47
DSP Software Considerations When Interfacing to the
AD73322L ....................................................................................37
Operating Mode ..........................................................................37
REVISION HISTORY
12/04—Rev. 0 to Rev. A
Updated Format.................................................................. Universal
Updated Outline Dimensions........................................................46
Changes to Ordering Guide...........................................................47
4/01—Revision 0: Initial Version
Rev. A | Page 3 of 48
AD73322L
SPECIFICATIONS
AVDD = 3 V 10ꢀ; DVDD = 3 V 10ꢀ; DGND = AGND = 0 V, fDMCLK = 16.384 MHz, fSAMP = 8 kHz; TA = TMIN to TMAX
,
unless otherwise noted.
Operating temperature range as follows: A grade, TMIN = −40°C, TMAX = +85°C; Y grade, TMIN = −40°C, TMAX = +105°C.
Table 1.
A and Y Versions
Parameter
REFERENCE
REFCAP
Min
Typ
Max
Unit
Test Conditions/Comments
Absolute Voltage, VREFCAP
REFCAP TC
REFOUT
1.08
1.2
50
1.32
V
ppm/°C 0.1 µF capacitor required from REFCAP to AGND2
Typical Output Impedance
Absolute Voltage, VREFOUT
Minimum Load Resistance
Maximum Load Capacitance
130
1.2
1
Ω
V
kΩ
pF
1.08
1.32
Unloaded
100
INPUT AMPLIFIER
Offset
1.0
1.578
50
mV
V
kΩ
pF
Maximum Output Swing
Feedback Resistance
Feedback Capacitance
ANALOG GAIN TAP
Gain at Maximum Setting
Gain at Minimum Setting
Gain Resolution
Max output swing = (1.578/1.2) × VREFCAP
fC = 32 kHz
100
+1
−1
5
1.0
1.0
0.5
Bits
%
µs
Gain step size = 0.0625
Output unloaded
Gain Accuracy
Settling Time
Tap gain change of −FS to +FS
Delay
µs
ADC SPECIFICATIONS
Maximum Input Range at VIN1, 2
DAC unloaded
1.578
−2.85
1.0954
−6.02
V p-p
dBm
V p-p
dBm
Measured differentially
Max input = (1.578/1.2) × VREFCAP
Measured differentially
Nominal Reference Level at VIN
(0 dBm0)
Absolute Gain
PGA = 0 dB
−2.0
70
−0.7
0.1
+0.5
−75
dB
dB
1.0 kHz, 0 dBm0
1.0 kHz, +3 dBm0 to −50 dBm0
Refer to Figure 9
300 Hz to 3400 Hz; fSAMP = 8 kHz, PUIA = 0
300 Hz to 3400 Hz; fSAMP = 8 kHz, PUIA = 1
0 Hz to fSAMP/2; fSAMP = 8 kHz
Gain Tracking Error
Signal-to-Noise and Distortion
PGA = 0 dB
78
79
77.5
dB
dB
dB
Total Harmonic Distortion
PGA = 0 dB
Intermodulation Distortion
Idle Channel Noise Crosstalk
ADC-to-DAC
−86
−61
−72
−107
dB
dB
dBm0
dB
300 Hz to 3400 Hz; fSAMP = 8 kHz
PGA = 0 dB
PGA = 0 dB
ADC input signal level: 1.0 kHz, 0 dBm0
DAC input at idle
ADC-to-ADC
−92
dB
ADC1 input signal level: 1.0 kHz, 0 dBm0
ADC2 input at idle; input amplifiers bypassed
Input amplifiers included in input channel
PGA = 0 dB
Input signal level at AVDD and DVDD pins:
1.0 kHz, 100 mV p-p sine wave
−93
0
−65
dB
mV
dB
DC Offset
Power Supply Rejection Ratio
−20
+20
Rev. A | Page 4 of 48
AD73322L
A and Y Versions
Parameter
Min
Typ
25
20
Max
Unit
µs
kΩ
Test Conditions/Comments
Group Delay3, 4
Input Resistance at PGA1, 3, 5
Input amplifiers bypassed
DIGITAL GAIN TAP
Gain at Maximum Setting
Gain at Minimum Setting
Gain Resolution
Delay
1
−1
16
25
100
Bits
µs
µs
Tested to 5 MSB of settings
Includes DAC delay
Tap gain change from −FS to +FS; includes
DAC settling time
Settling Time
DAC SPECIFICATIONS
Maximum Voltage Output Swing1
Single-Ended
DAC unloaded
1.578
−2.85
3.156
3.17
V p-p
dBm
V p-p
dBm
PGA = 6 dB
Max output = (1.578/1.2) × VREFCAP
PGA = 6 dB
Differential
Max output = 2 × (1.578/1.2) × VREFCAP
Nominal Voltage Output Swing (0 dBm0)
Single-Ended
1.0954
−6.02
2.1909
0
V p-p
dBm
V p-p
dBm
V
PGA = 6 dB
PGA = 6 dB
Differential
Output Bias Voltage
1.2
REFOUT unloaded
Absolute Gain
Gain Tracking Error
Signal-to-Noise and Distortion at 0 dBm0
PGA = 0 dB
−1.75
72
−0.6
0.1
+0.75 dB
dB
1.0 kHz, 0 dBm0; unloaded
1.0 kHz, +3 dBm0 to −50 dBm0
Refer to Figure 10
78.5
dB
300 Hz to 3400 Hz; fSAMP = 8 kHz
Total Harmonic Distortion at 0 dBm0
PGA = 0 dB
Intermodulation Distortion
Idle Channel Noise Crosstalk
DAC-to-ADC
−89
−77
−81
−73
−75
dB
dB
dBm0
dB
300 Hz to 3400 Hz; fSAMP = 8 kHz
PGA = 0 dB
PGA = 0 dB
ADC input signal level: AGND;
DAC output signal level: 1.0 kHz, 0 dBm0
Input amplifiers bypassed
−74
−102
dB
dB
Input amplifiers included in input channel
DAC1 output signal level: AGND; DAC2
Output signal level: 1.0 kHz, 0 dBm0
DAC-to-DAC
Power Supply Rejection
Group Delay3, 4
−65
dB
Input signal level at AVDD and DVDD pins:
1.0 kHz, 100 mV p-p sine wave
Interpolator bypassed
25
50
+5
µs
µs
mV
Output DC Offset1, 6
Minimum Load Resistance, RL
Single-Ended3
Differential
Maximum Load Capacitance, CL
−50
+60
1, 7
150
150
Ω
Ω
1, 7
Single-Ended
Differential
500
100
pF
pF
FREQUENCY RESPONSE
(ADC and DAC)8 Typical Output
Frequency (Normalized to FS)
0
0
−0.1
dB
dB
0.03125
Rev. A | Page 5 of 48
AD73322L
A and Y Versions
Typ
Parameter
Min
Max
Unit
dB
dB
dB
dB
dB
dB
dB
dB
Test Conditions/Comments
0.0625
0.125
0.1875
0.25
0.3125
0.375
0.4375
> 0.5
−0.25
−0.6
−1.4
−2.8
− 4.5
−7.0
−9.5
<−12.5
LOGIC INPUTS
VINH, Input High Voltage
VINL, Input Low Voltage
IIH, Input Current
CIN, Input Capacitance
LOGIC OUTPUT
VOH, Output High Voltage
VOL, Output Low Voltage
Three-State Leakage Current
POWER SUPPLIES
AVDD1, AVDD2
DVDD
0.8
+10
10
V
DVDD − 0.8
0
−10
V
µA
pF
DVDD
0.4
+10
V
|IOUT| ≤100 µA
|IOUT| ≤100 µA
DVDD − 0.4
0
−10
V
µA
2.7
2.7
3.3
3.3
V
V
9
DVDD IDD
See Table 2
1 Test conditions: input PGA set for 0 dB gain, output PGA set for 6 dB gain, no load on analog outputs (unless otherwise noted).
2 At input to sigma-delta modulator of ADC.
3 Guaranteed by design.
4 Overall group delay is affected by the sample rate and the external digital filtering.
5 The ADC’s input impedance is inversely proportional to DMCLK and is approximated by (3/3 × 1011)/DMCLK.
6 Between VOUTP1 and VOUTN1 or between VOUTP2 and VOUTN2.
7 At VOUT output.
8 Frequency responses of ADC and DAC measured with input at audio reference level (the input level that produces an output level of −10 dBm0), with 38 dB
preamplifier bypassed and input gain of 0 dB.
9 Test conditions: no load on digital inputs, analog inputs ac-coupled to ground, no load on analog outputs.
CURRENT SUMMARY
AVDD = DVDD = 3.3 V. These values are in mA and are typical values unless otherwise noted.
Table 2.
Analog
Current
Digital
Current
Total
Current (Typ)
Total
Current (Max)
Conditions
SE
1
1
1
1
MCLK ON
YES
YES
YES
YES
Comments
ADCs on only
DACs on only
ADCs and DACs on
ADCs and DACs and
Input amps on
3.4
8.8
11.6
13.8
6.3
6.5
7.0
7.0
9.7
12
20
23
26
REFOUT disabled
REFOUT disabled
REFOUT disabled
REFOUT disabled
15.3
18.6
20.8
ADCs and DACs and
AGT on
All sections on
REFCAP on only
REFCAP and REFOUT
On only
13.2
7.0
20.2
26
1
YES
REFOUT disabled
REFOUT disabled
17.2
0.65
2.56
7.0
0
0
24.2
0.67
2.57
31
1.25
4.5
1
0
0
YES
NO
NO
All sections off
0
1.25
1.25
1.8
0
0
YES
NO
MCLK active levels equal
to 0 V and DVDD
Digital inputs static and
Equal to 0 V or DVDD
All sections off
0 µA
12.5 µA
12.7 µA
40 µA
Rev. A | Page 6 of 48
AD73322L
SIGNAL RANGES
Table 3.
Mnemoic
VREFCAP
VREFOUT
ADC
Description
Range
1.2 V 10%
1.2 V 10%
1.578 V p-p
1.0954 V p-p
Maximum input range at VIN
Nominal reference level
Maximum voltage output swing
Single-Ended
DAC
1.578 V p-p
3.156 V p-p
Differential
Nominal voltage output swing
Single-Ended
Differential
1.0954 V p-p
2.1909 V p-p
VREFOUT
Output bias voltage
TIMING CHARACTERISTICS
AVDD = 3 V 10ꢀ; DVDD = 3 V 10ꢀ; AGND = DGND = 0 V; TA = TMlN to TMAX, unless otherwise noted.
Table 4.
Parameter
Unit
Description
Limit at TA = −40°C to +105°C
Clock Signals
See Figure 2
t1
t2
t3
61
24.4
24.4
ns min
ns min
ns min
MCLK period
MCLK width high
MCLK width low
Serial Port
t4
t5
t6
t7
t8
t9
t10
t11
t12
t13
See Figure 4 and Figure 5
SCLK period
SCLK width high
t1
ns min
ns min
ns min
ns min
ns min
ns max
ns min
ns min
ns max
ns max
0.4 × t1
0.4 × t1
20
0
10
10
10
10
30
SCLK width low
SDI/SDIFS setup before SCLK low
SDI/SDIFS hold after SCLK low
SDOFS delay from SCLK high
SDOFS hold after SCLK high
SDO hold after SCLK high
SDO delay from SCLK high
SCLK delay from MCLK
Rev. A | Page 7 of 48
AD73322L
TIMING DIAGRAMS
t1
t2
t3
Figure 2. MCLK Timing
100µA
I
OL
TO OUTPUT
PIN
2.1V
C
L
15pF
100µA
I
OH
Figure 3. Load Circuit for Timing Specifications
t1
t2
t3
MCLK
SCLK*
t13
t5
t6
t4
* SCLK IS INDIVIDUALLY PROGRAMMABLE
IN FREQUENCY (MCLK/4 SHOWN HERE).
Figure 4. SCLK Timing
SE (I)
THREE-
STATE
SCLK (O)
t7
SDIFS (I)
t8
t8
t7
SDI (I)
D15
D14
D1
D0
D15
t9
t10
THREE-
STATE
SDOFS (O)
t12
t11
THREE-
STATE
SDO (O)
D15
D2
D1
D0
D15
D14
Figure 5. Serial Port (SPORT)
Rev. A | Page 8 of 48
AD73322L
ABSOLUTE MAXIMUM RATINGS
TA = 25°C unless otherwise noted.
Table 5.
Parameters
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.
Ratings
AVDD, DVDD to GND
AGND to DGND
Digital I/O Voltage to DGND
Analog I/O Voltage to AGND
Operating Temperature Range
Industrial (A Version)
−0.3 V to +4.6 V
−0.3 V to +0.3 V
−0.3 V to (DVDD + 0.3 V)
−0.3 V to (AVDD + 0.3 V)
−40°C to +85°C
−40°C to +105°C
−65°C to +150°C
Extended (Y Version)
Storage Temperature Range
Maximum Junction Temperature 150°C
SOIC, θJA Thermal Impedance
Lead Temperature, Soldering
Vapor Phase (60 sec)
71.4°C/W
215°C
Infrared (15 sec)
220°C
LQFP, θJA Thermal Impedance
Lead Temperature, Soldering
Vapor Phase (60 sec)
53.2°C/W
215°C
Infrared (15 sec)
220°C
TSSOP, θJA Thermal Impedance
Lead Temperature, Soldering
Vapor Phase (60 sec)
97.9°C/W
215°C
220°C
Infrared (15 sec)
ESD CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily accumulate on
the uman 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. A | Page 9 of 48
AD73322L
PIN CONFIGURATIONS AND FUNCTION DESCRIPTIONS
VINP1
VFBP1
VINN1
1
2
3
4
5
6
7
8
9
28 VFBN2
27 VINN2
26 VFBP2
25 VINP2
24 VOUTN1
23 VOUTP1
22 VOUTN2
21 VOUTP2
20 AVDD1
19 AGND1
18 SE
1
2
28
27
26
25
24
23
22
21
20
19
18
17
16
15
VINP1
VFBP1
VINN1
VFBN2
VINN2
VFBP2
VINP2
3
VFBN1
REFOUT
REFCAP
AVDD2
AGND2
DGND
4
VFBN1
REFOUT
REFCAP
AVDD2
AGND2
DGND
AD73322L
TOP VIEW
(Not to Scale)
5
AD73322L
TOP VIEW
VOUTN1
VOUTP1
VOUTN2
VOUTP2
AVDD1
AGND1
SE
6
7
8
9
10
11
12
13
14
DVDD 10
RESET 11
SCLK 12
MCLK 13
SDO 14
DVDD
RESET
SCLK
SDI
17 SDI
MCLK
SDIFS
SDOFS
16 SDIFS
15 SDOFS
SDO
Figure 6. 28-Lead Wide Body
Figure 7. 28-Lead Thin Shrink
44 43 42 41 40 39 38 37 36 35 34
1
2
33
32
31
30
29
28
27
26
25
24
23
REFOUT
NC
PIN 1
REFCAP
AVDD2
AVDD2
AGND2
AGND2
AGND2
AGND2
DGND
VOUTN1
VOUTP1
NC
3
4
5
VOUTN2
VOUTP2
NC
AD73322L
TOP VIEW
(Not to Scale)
6
7
8
AVDD1
AVDD1
AGND1
AGND1
9
10
11
DGND
DVDD
12 13 14 15 16 17 18 19 20 21 22
NC = NO CONNECT
Figure 8. 44-Lead Plastic Thin Quad Flatpack
Table 6. Pin Function Descriptions
Mnemonic Function
VINP1
VFBP1
Analog Input to the inverting input amplifier on Channel 1’s positive input.
Feedback Connection from the output of the inverting amplifier on Channel 1’s positive input. When the input amplifiers are
bypassed, this pin allows direct access to the positive input of Channel 1’s sigma-delta modulator.
VINN1
VFBN1
Analog Input to the inverting input amplifier on Channel 1’s negative input.
Feedback connection from the output of the inverting amplifier on Channel 1’s negative input. When the input amplifiers are
bypassed, this pin allows direct access to the negative input of Channel 1’s sigma-delta modulator.
REFOUT
REFCAP
AVDD2
AGND2
DGND
Buffered Reference Output, which has a nominal value of 1.2 V.
A bypass capacitor to AGND2 of 0.1 µF is required for the on-chip reference. The capacitor should be fixed to this pin.
Analog Power Supply Connection.
Analog Ground/Substrate Connection2.
Digital Ground/Substrate Connection.
DVDD
Digital Power Supply Connection.
RESET
Active Low Reset Signal. This input resets the entire chip, resetting the control registers and clearing the digital circuitry.
Rev. A | Page 10 of 48
AD73322L
Mnemonic Function
SCLK
Serial Clock Output. This rate determines the serial transfer rate to/from the codec. It is used to clock data or control
information to and from the serial port (SPORT). The frequency of SCLK is equal to the frequency of the master clock (MCLK)
divided by an integer number—this integer number being the product of the external master clock rate divider and the serial
clock rate divider.
MCLK
SDO
Master Clock Input. MCLK is driven from an external clock signal.
Serial Data Output. Both data and control information may be output on this pin and are clocked on the positive edge of
SCLK. SDO is in three-state when no information is being transmitted and when SE is low.
SDOFS
SDIFS
SDI
Framing Signal Output for SDO Serial Transfers. The frame sync is one bit wide and is active one SCLK period before the first
bit (MSB) of each output word. SDOFS is referenced to the positive edge of SCLK. SDOFS is in three-state when SE is low.
Framing Signal Input for SDI Serial Transfers. The frame sync is one bit wide and is valid one SCLK period before the first bit
(MSB) of each input word. SDIFS is sampled on the negative edge of SCLK and is ignored when SE is low.
Serial Data Input. Both data and control information may be input on this pin and are clocked on the negative edge of SCLK.
SDI is ignored when SE is low.
SE
SPORT Enable. Asynchronous input enable pin for the SPORT. When SE is set low by the DSP, the output pins of the SPORT are
three-stated and the input pins are ignored. SCLK is also disabled internally in order to decrease power dissipation. When SE is
brought high, the control and data registers of the SPORT are at their original values (before SE was brought low); however,
the timing counters and other internal registers are at their reset values.
AGND1
AVDD1
VOUTP2
VOUTN2
VOUTP1
VOUTN1
VINP2
Analog Ground/Substrate Connection.
Analog Power Supply Connection.
Analog Output from the Positive Terminal of Output Channel 2.
Analog Output from the Negative Terminal of Output Channel 2.
Analog Output from the Positive Terminal of Output Channel 1.
Analog Output from the Negative Terminal of Output Channel 1.
Analog Input to the inverting input amplifier on Channel 2’s positive input.
VFBP2
Feedback connection from the output of the inverting amplifier on Channel 2’s positive input. When the input amplifiers are
bypassed, this pin allows direct access to the positive input of Channel 2’s sigma-delta modulator.
VINN2
VFBN2
Analog Input to the inverting input amplifier on Channel 2’s negative input.
Feedback connection from the output of the inverting amplifier on Channel 2’s negative input. When the input amplifiers are
bypassed, this pin allows direct access to the negative input of Channel 2’s sigma-delta modulator.
Rev. A | Page 11 of 48
AD73322L
TERMINOLOGY
Absolute Gain
Sample Rate
A measure of converter gain for a known signal. Absolute gain
is measured (differentially) with a 1 kHz sine wave at 0 dBm0
for the DAC and with a 1 kHz sine wave at 0 dBm0 for the
ADC. The absolute gain specification is used for gain tracking
error specification.
The rate at which the ADC updates its output register and the
DAC updates its output from its input register. The sample rate
can be chosen from a list of four that are fixed relative to the
DMCLK. Sample rate is set by programming bits DIR0-1 in
Control Register B of each channel.
Crosstalk
SNR + THD
Crosstalk is due to coupling of signals from a given channel to
an adjacent channel. It is defined as the ratio of the amplitude
of the coupled signal to the amplitude of the input signal.
Crosstalk is expressed in dB.
Signal-to-noise ratio plus harmonic distortion is the ratio of the
rms value of the measured input signal to the rms sum of all
other spectral components in the frequency range 300 Hz to
3400 Hz, including harmonics but excluding dc.
Gain Tracking Error
ABBREVIATIONS
Measures changes in converter output for different signal levels
relative to an absolute signal level. The absolute signal level is
0 dBm0 (equal to absolute gain) at 1 kHz for the DAC and 0
dBm0 (equal to absolute gain) at 1 kHz for the ADC. Gain
tracking error at 0 dBm0 (ADC) and 0 dBm0 (DAC) is 0 dB
by definition.
Table 7.
Abbreviation Definition
ADC
AFE
AGT
ALB
BW
Analog-to-digital converter.
Analog front end.
Analog gain tap.
Analog loop-back.
Bandwidth.
A control register where x is a placeholder for
an alphabetic character (A to H). There are eight
read/write control registers on the AD73322L—
CRA through CRH.
A bit position, where n is a placeholder for a
numeric character (0 to 7), within a control
register, where x is a placeholder for an
alphabetic character (A to E). Position 7
represents the MSB and Position 0 represents
the LSB.
Group Delay
CRx
The derivative of radian phase with respect to radian frequency,
dø(f)/df. Group delay is a measure of the average delay of a
system as a function of frequency. A linear system with a
constant group delay has a linear phase response. The deviation
of group delay from a constant indicates the degree of nonlinear
phase response of the system.
CRx:n
Idle Channel Noise
The total signal energy measured at the output of the device
when the input is grounded (measured in the frequency range
300 Hz to 3400 Hz).
DAC
DGT
DLB
Digital-to-analog converter.
Digital gain tap.
Digital loop-back.
Intermodulation Distortion
DMCLK
Device (internal) master clock. This is the
internal master clock resulting from the
external master clock (MCLK) being divided by
the on-chip master clock divider.
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, etc. Intermodulation terms are those for which
neither m nor n is equal to zero. For final testing, 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).
FS
FSLB
Full scale.
Frame sync loop-back—where the SDOFS of
the final device in a cascade is connected to the
RFS and TFS of the DSP and the SDIFS of first
device in the cascade. Data input and output
occur simultaneously. In the case of nonFSLB,
SDOFS and SDO are connected to the Rx port of
the DSP while SDIFS and SDI are connected to
the Tx port.
Programmable gain amplifier.
Switched capacitor.
SPORT loop-back.
Power Supply Rejection
Measures the susceptibility of a device to noise on the power
supply. Power supply rejection is measured by modulating the
power supply with a sine wave and measuring the noise at the
output (relative to 0 dB).
PGA
SC
SLB
SNR
SPORT
THD
VBW
Signal-to-noise ratio.
Serial port.
Total harmonic distortion.
Voice bandwidth.
Rev. A | Page 12 of 48
AD73322L
TYPICAL PERFORMANCE CHARACTERISTICS AND FUNCTIONAL BLOCK DIAGRAM
80
70
60
50
40
30
20
10
0
80
70
60
50
40
30
20
10
0
–10
–85
–10
–85
–75
–65
–55
–45
–35
–25
–15
–5
5
–75
–65
–55
–45
–35
–25
–15
–5
5
3.17
3.17
V
(dBm0)
V
(dBm0)
IN
IN
Figure 9. S/N(N = D) vs. VIN (ADC @ 3 V) over Voice Bandwidth
(300 Hz to 3.4 kHz)
Figure 10. S/N(N = D) vs. VIN (DAC @ 3 V) over Voice Bandwidth
(300 Hz to 3.4 kHz)
AVDD1
AVDD2
DVDD
VFBN1
VINN1
ANALOG
LOOP-
BACK
INVERT
SINGLE-ENDED
ENABLE
DIGITAL
Σ-∆
MODULATOR
0/38dB
PGA
V
DECIMATOR
REF
SDI
SDIFS
SCLK
VINP1
VFBP1
GAIN
±1
GAIN
±1
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
SWITCHED
CAPACITOR
LOW-PASS
FILTER
DIGITAL
Σ-∆
MODULATOR
1-BIT
DAC
INTER-
POLATOR
+6/15dB
PGA
RESET
MCLK
SE
SERIAL
I/O
PORT
REFCAP
REFOUT
REFERENCE
AD73322L
VFBN2
VINN2
SDO
ANALOG
LOOP-
BACK
INVERT
SINGLE-ENDED
ENABLE
DIGITAL
Σ-∆
MODULATOR
SDOFS
0/38dB
PGA
V
DECIMATOR
REF
VINP2
VFBP2
GAIN
±1
GAIN
±1
VOUTP2
VOUTN2
CONTINUOUS
TIME
LOW-PASS
FILTER
SWITCHED
CAPACITOR
LOW-PASS
FILTER
DIGITAL
Σ-∆
MODULATOR
1-BIT
DAC
INTER-
POLATOR
+6/–15dB
PGA
AGND1
AGND2
DGND
Figure 11. Functional Block Diagram
Rev. A | Page 13 of 48
AD73322L
FUNCTIONAL DESCRIPTIONS
ENCODER CHANNELS
highest frequency of interest. In the case of the AD73322L, the
initial sampling rate of the sigma-delta modulator is DMCLK/8.
The main effect of oversampling is that the quantization noise is
spread over a very wide bandwidth, up to FS/2 = DMCLK/16
(Figure 13). This means that the noise in the band of interest is
much reduced. Another complementary feature of sigma-delta
converters is the use of a technique called noise-shaping. This
technique has the effect of pushing the noise from the band of
interest to an out-of-band position (Figure 14). The combin-
ation of these techniques, followed by the application of a
digital filter, sufficiently reduces the noise in band to ensure
good dynamic performance from the part (Figure 15).
Both encoder channels consist of a pair of inverting op amps
with feedback connections that can be bypassed if required, a
switched capacitor PGA and a sigma-delta analog-to-digital
converter (ADC). An on-board digital filter, which forms part
of the sigma-delta ADC, also performs critical system-level
filtering. Due to the high level of oversampling, the input
antialias requirements are reduced such that a simple single-
pole RC stage is sufficient to give adequate attenuation in the
band of interest.
PROGRAMMABLE GAIN AMPLIFIER
Each encoder section’s analog front end comprises a switched
capacitor PGA, which also forms part of the sigma-delta
modulator. The SC sampling frequency is DMCLK/8. The
PGA, whose programmable gain settings are shown in Table 8,
may be used to increase the signal level applied to the ADC
from low output sources such as microphones, and can be
used to avoid placing external amplifiers in the circuit. The
input signal level to the sigma-delta modulator should not
exceed the maximum input voltage permitted.
BAND
OF
F /2
S
DMCLK/16
INTEREST
A.
The PGA gain is set by bits IGS0, IGS1, and IGS2 (CRD:0–2) in
control register D.
Table 8. PGA Settings for the Encoder Channel
NOISE SHAPING
IGS2
IGS1
IGS0
Gain (dB)
BAND
OF
INTEREST
F /2
S
DMCLK/16
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
0
6
B.
12
18
20
26
32
38
DIGITAL FILTER
BAND
OF
F /2
S
DMCLK/16
INTEREST
ADC
C.
Both ADCs consist of an analog sigma-delta modulator and a
digital antialiasing decimation filter. The sigma-delta modulator
noise-shapes the signal and produces 1-bit samples at a
DMCLK/8 rate. This bit stream, representing the analog input
signal, is input to the antialiasing decimation filter. The
decimation filter reduces the sample rate and increases the
resolution.
Figure 12. Sigma-Delta Noise Reduction
Figure 13 through Figure 16 show the various stages of filtering
that are employed in a typical AD73322L application. Figure 13
shows the transfer function of the external analog antialias
filter. Even though it is a single RC pole, its cutoff frequency
is sufficiently far away from the initial sampling frequency
(DMCLK/8) that it takes care of any signals that could be
aliased by the sampling frequency. This also shows the major
difference between the initial oversampling rate and the band-
width of interest. In Figure 14, the signal and noise-shaping
responses of the sigma-delta modulator are shown. The
signal response provides further rejection of any high
ANALOG SIGMA-DELTA MODULATOR
The AD73322L’s input channels employ a sigma-delta
conversion technique, which provides a high resolution 16-bit
output with system filtering being implemented on-chip.
Sigma-delta converters employ a technique known as
oversampling, where the sampling rate is many times the
frequency signals, while the noise-shaping pushes the inherent
quantization noise to an out-of-band position. The detail of
Rev. A | Page 14 of 48
AD73322L
Figure 15 shows the response of the digital decimation filter
(sinc-cubed response) with nulls every multiple of DMCLK/256
corresponding to the decimation filter update rate for a 64 kHz
sampling. The nulls of the Sinc3 response correspond with
multiples of the chosen sampling frequency. The final detail in
Figure 16 shows the application of a final antialias filter in the
DSP engine. This has the advantage of being implemented
according to the user’s requirements and available MIPS. The
filtering in Figure 13 through Figure 16 is implemented in the
AD73322L.
The antialiasing decimation filter is a sinc-cubed digital filter
that reduces the sampling rate from DMCLK/8 to DMCLK/256,
and increases the resolution from a single bit to 15 bits or
greater (depending on chosen sampling rate). Its Z transform is
given as
[(1 − Z−N )/(1 − Z−1 )]3
where N is set by the sampling rate
(N = 32 @ 64 kHz sampling N = 256 @ 8 kHz sampling)
Thus, when the sampling rate is 64 kHz, a minimal group delay
of 25 µs can be achieved.
Figure 13 to Figure 16 show ADC frequency responses.
Word growth in the decimator is determined by the sampling
rate. At 64 kHz sampling, where the oversampling ratio (OSR)
between sigma-delta modulator and decimator output equals
32, there are five bits per stage of the three-stage Sinc3 filter.
Due to symmetry within the sigma-delta modulator, the LSB
is always a zero; therefore, the 16-bit ADC output word has
2 LSBs equal to zero, one due to the sigma-delta symmetry and
the other being a padding zero to make up the 16-bit word. At
lower sampling rates, decimator word growth is greater than the
16-bit sample word, therefore truncation occurs in transferring
the decimator output as the ADC word. For example, at 8 kHz
sampling, word growth reaches 24 bits due to the OSR of 256
between the sigma-delta modulator and decimator output. This
yields 8 bits per stage of the three-stage sinc3 filter.
F
B
= 4kHz
F
= DMCLK/8
SINIT
Figure 13. Analog Antialias Filter Transfer Function
SIGNAL TRANSFER FUNCTION
NOISE TRANSFER FUNCTION
ADC CODING
F
B
= 4kHz
F
= DMCLK/8
The ADC coding scheme is in twos complement format, as
shown in Figure 17). The output words are formed by the
decimation filter, which grows the word length from the single
bit output of the sigma-delta modulator to a word length of up
to 24 bits (depending on decimation rate chosen), which is the
final output of the ADC block. In data mode this value is trun-
cated to 16 bits for output on the serial data output (SDO) pin.
SINIT
Figure 14. Analog Sigma-Delta Modulator Transfer Function
V
+ (V
× 0.32875)
REF
V
REF
INN
F
= 4kHz
F
= DMCLK/256
B
SINTER
ANALOG
INPUT
V
REF
Figure 15. Digital Decimator Transfer Function
V
V
– (V
× 0.32875)
REF
INP
REF
10...00
00...00
01...11
ADC CODE DIFFERENTIAL
V
+ (V
× 0.6575)
REF
REF
F
= 4kHz
F
= 8kHz
F
= DMCLK/256
B
SRNAL
SINTER
V
INN
Figure 16. Final Filter (HPF) Transfer Function
ANALOG
INPUT
DECIMATION FILTER
The digital filter used in the AD73322L carries out two
important functions. First, it removes the out-of-band
quantization noise, which is shaped by the analog modulator
and second, it decimates the high frequency bit stream to a
lower rate, 16-bit word.
V
INP
V
– (V
× 0.6575)
REF
REF
10...00
00...00
01...11
ADC CODE SINGLE-ENDED
Figure 17. ADC Transfer Function
Rev. A | Page 15 of 48
AD73322L
In mixed control/data mode, the resolution is fixed at 15 bits,
with the MSB of the 16-bit transfer being used as a flag bit to
indicate either control or data in the frame.
Table 9. PGA Settings for the Decoder Channel
OGS2
OGS1
OGS0
Gain (dB)
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
+6
+3
0
DECODER CHANNEL
The decoder channels consist of digital interpolators, digital
sigma-delta modulators, single bit digital-to-analog converters
(DAC), analog smoothing filters and programmable gain
amplifiers with differential outputs.
−3
−6
−9
−12
−15
DAC CODING
The DAC coding scheme is in twos complement format with
0x7FFF being full-scale positive and 0x8000 being full-scale
negative.
DIFFERENTIAL OUTPUT AMPLIFIERS
The decoder has a differential analog output pair (VOUTP and
VOUTN). The output channel can be muted by setting the
MUTE bit (CRD:7) in Control Register D. The output signal is
dc-biased to the codec’s on-chip voltage reference.
INTERPOLATION FILTER
The anti-imaging interpolation filter is a sinc-cubed digital
filter that up-samples the 16-bit input words from the input
sample rate to a rate of DMCLK/8, while filtering to attenuate
images produced by the interpolation process. Its Z transform is
given as
VOLTAGE REFERENCE
The AD73322L reference, REFCAP, is a band gap reference that
provides a low noise, temperature-compensated reference to the
DAC and ADC. A buffered version of the reference is also made
available on the REFOUT pin, and can be used to bias other
external analog circuitry. The reference has a default nominal
value of 1.2 V.
[(1 − Z−N )/(1 − Z−1 )]3
where N is determined by the sampling rate
(N = 32 @ 64 kHz . . . N = 256 @ 8 kHz)
The DAC receives 16-bit samples from the host DSP processor
at the programmed sample rate of DMCLK/N. If the host
processor fails to write a new value to the serial port, the
existing (previous) data is read again. The data stream is filtered
by the anti-imaging interpolation filter, but there is an option to
bypass the interpolator for the minimum group delay
configuration by setting the IBYP bit (CRE:5) of Control
Register E. The interpolation filter has the same characteristics
as the ADC’s antialiasing decimation filter.
The reference output (REFOUT) can be enabled for biasing
external circuitry by setting the RU bit (CRC:6) of CRC.
ANALOG
LOOP-BACK
SELECT
SINGLE-
ENDED
ENABLE
INVERTING
OP AMPS
INVERT
VFBN1
VINN1
The output of the interpolation filter is fed to the DAC’s digital
sigma-delta modulator, which converts the 16-bit data to 1-bit
samples at a rate of DMCLK/8. The modulator noise-shapes the
signal so that errors inherent to the process are minimized in
the pass band of the converter. The bit-stream output of the
sigma-delta modulator is fed to the single bit DAC where it is
converted to an analog voltage.
0/38dB
PGA
V
REF
VINP1
VFBP1
V
REF
GAIN
±1
ANALOG
GAIN TAP
ANALOG SMOOTHING FILTER AND PGA
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
The output of the single bit DAC is sampled at DMCLK/8,
therefore it is necessary to filter the output to reconstruct the
low frequency signal. The decoder’s analog smoothing filter
consists of a continuous-time filter preceded by a third-order
switched-capacitor filter. The continuous-time filter forms
part of the output programmable gain amplifier (PGA).
REFCAP
REFOUT
REFERENCE
AD73322L
The PGA can be used to adjust the output signal level from
−15 dB to +6 dB in 3 dB steps, as shown in Table 9. The PGA
gain is set by bits OGS0, OGS1, and OGS2 (CRD:4-6) in
Control Register D.
Figure 18. Analog Input/Output Section
Rev. A | Page 16 of 48
AD73322L
MCLK
MCLK
EXTERNAL
EXTERNAL
DMCLK INTERNAL
DMCLK INTERNAL
MCLK
MCLK
DIVIDER
DIVIDER
SCLK
SCLK
3
SCLK
3
SCLK
DIVIDER
DIVIDER
SE
RESET
SDIFS
SDI
SE
SERIAL PORT 1
(SPORT 1)
SERIAL PORT 2
(SPORT 1)
RESET
SDOFS
SDO
SDOFS1
SDO1
SDIFS2
SDI2
SERIAL REGISTER 1
SERIAL REGISTER 2
2
2
8
8
8
8
8
CONTROL
REGISTER
1E
8
CONTROL
REGISTER
2E
8
CONTROL
REGISTER
1B
8
8
CONTROL
REGISTER
2B
8
CONTROL
REGISTER
1A
CONTROL
REGISTER
1C
CONTROL
REGISTER
1D
CONTROL
REGISTER
2A
CONTROL
REGISTER
2C
CONTROL
REGISTER
2D
16
8
16
8
CONTROL
REGISTER
1G
CONTROL
REGISTER
1F
CONTROL
REGISTER
2G
CONTROL
REGISTER
2F
CONTROL
REGISTER
1H
CONTROL
REGISTER
2H
Figure 19. SPORT Block Diagram
ANALOG AND DIGITAL GAIN TAPS
The AD73322L features analog and digital feedback paths
between input and output. The amount of feedback is deter-
mined by the gain setting which is programmed in the control
registers. This feature can typically be used for balancing the
effective impedance between input and output when used in
subscriber line interface circuit (SLIC) interfacing.
Control bit CRF:5 connects/disconnects the AGT output to the
summer block at the output of the DAC section while control
bit CRF:7 overrides the gain tap setting with a mute, (zero gain)
setting. Table 10 shows the gain vs. digital setting for the AGT.
In this table, AGT and DGT weights are given for the case of
VFBNx (connected to the sigma-delta modulator’s positive
input) being at a higher potential than VFBPx (connected to the
sigma-delta modulator’s negative input).
Analog Gain Tap
The analog gain tap is configured as a programmable
differential amplifier whose input is taken from the ADC’s
input signal path. The output of the analog gain tap is summed
with the output of the DAC. The gain is programmable using
Control Register F (CRF:0-4) to achieve a gain of −1 to +1 in
32 steps with muting being achieved through a separate control
setting (Control Register F Bit 7). The gain increment per step
is 0.0625. The AGT is enabled by powering-up the AGT control
bit in the power control register (CRC:1). When this bit is
set (=1), CRF becomes an AGT control register with CRF:0-4
holding the AGT coefficient, CRF:5 becomes an AGT enable
and CRF:7 becomes an AGT mute control bit.
Table 10. Analog Gain Tap Settings
AGTC4
AGTC3
AGTC2
AGTC1
AGTC0
Gain (dB)
1.00
0.9375
0.875
0.8125
0.75
0.0625
−0.0625
−0.875
−0.9375
−1.00
0
0
0
0
0
0
1
1
1
1
0
0
0
0
0
1
0
1
1
1
0
0
0
0
1
1
0
1
1
1
0
0
1
1
0
1
0
0
1
1
0
1
0
1
0
1
0
1
0
1
Rev. A | Page 17 of 48
AD73322L
DIGITAL GAIN TAP
SPORT OVERVIEW
The digital gain tap features a programmable gain block whose
input is taken from the bit stream output of the ADC’s sigma
delta modulator. This single bit input (1 or 0) is used to add or
subtract a programmable value, which is the digital gain tap
setting, to the output of the DAC section’s interpolator. The
programmable setting has 16-bit resolution and is programmed
using the settings in Control Registers G and H, as shown in
Table 11. In this table, AGT and DGT weights are given for the
case of VFBNx (connected to the sigma-delta modulator’s
positive input) being at a higher potential than VFBPx
The AD73322L SPORT is a flexible, full-duplex, synchronous
serial port having a protocol designed to allow up to four
AD73322L devices (or combinations of AD73322L dual
codecs and AD73311 single codecs up to eight codec blocks) to
be connected, in cascade, to a single DSP via a 6-wire interface.
It has a very flexible architecture that can be configured by
programming two of the internal control registers in each codec
block. The device has three distinct modes of operation: control
mode, data mode, and mixed control/data mode.
Note that because each codec has its own SPORT section, the
register settings in both SPORTs must be programmed. The
registers that control SPORT and sample rate operation
(CRA and CRB) must be programmed with the same values,
otherwise incorrect operation may occur.
(connected to the sigma-delta modulator’s negative input).
Table 11. Digital Gain Tap Settings
DGT15–0 (Hex)
Gain
0x8000
0x9000
0xA000
0xC000
0xE000
−1.00
−0.875
−0.75
−0.5
−0.25
0.00
In control mode (CRA:0 = 0), the device’s internal configuration
can be programmed by writing to the eight internal control
registers. In this mode, control information can be written to or
read from the codec. In data mode (CRA:0 = 1), (CRA:1 = 0),
information sent to the device is used to update the decoder
section (DAC), while the encoder section (ADC) data is read
from the device. In this mode, only DAC and ADC data are
written to or read from the device. Mixed mode (CRA:0 = 1
and CRA:1 = 1) allows the user to choose whether the infor-
mation being sent to the device contains control information
or DAC data. This is achieved by using the MSB of the 16-bit
frame as a flag bit. Mixed mode reduces the resolution to 15 bits
with the MSB being used to indicate whether the information in
the 16-bit frame is control information or DAC/ADC data.
0x0000
0x2000
0x4000
0x6000
0x7FFF
+0.25
+0.05
+0.75
+0.99999
SERIAL PORT (SPORT)
The codecs communicate with a host processor via the
bidirectional synchronous serial port (SPORT), which is
compatible with most modern DSPs. The SPORT is used to
transmit and receive digital data and control information. The
dual codec is implemented using two separate codec blocks that
are internally cascaded with serial port access to the input of
Codec 1 and the output of Codec 2. This allows other single or
dual codec devices to be cascaded together (up to a limit of
eight codec units).
The SPORT features a single 16-bit serial register that is used
for both input and output data transfers. As the input and
output data must share the same register, some precautions
must be observed. The primary precaution is that no informa-
tion must be written to the SPORT without reference to an
output sample event, which is when the serial register is
overwritten with the latest ADC sample word. Once the SPORT
starts to output the latest ADC word, it is safe for the DSP to
write new control or data-words to the codec. In certain con-
figurations, data can be written to the device to coincide with
the output sample being shifted out of the serial register — see
the Interfacing section. The serial clock rate (CRB:2–3) defines
how many 16-bit words can be written to a device before the
next output sample event happens.
In both transmit and receive modes, data is transferred at the
serial clock (SCLK) rate with the MSB being transferred first.
Due to the fact that the SPORT of each codec block uses a
common serial register for serial input and output, commun-
ications between an AD73322L codec and a host processor
(DSP engine) must always be initiated by the codecs themselves.
In this configuration, the codecs are described as being in
master mode. This ensures that there is no collision between
input data and output samples.
Rev. A | Page 18 of 48
AD73322L
The SPORT block diagram shown in Figure 19 details the
blocks associated with Codecs 1 and 2, including the eight
control registers (A–H), external MCLK to internal DMCLK
divider, and serial clock divider. The divider rates are controlled
by the setting of Control Register B. The AD73322L features a
master clock divider that allows users the flexibility of dividing
externally available high frequency DSP or CPU clocks to
generate a lower frequency master clock internally in the codec,
which may be more suitable for either serial transfer or
sampling rate requirements. The master clock divider has five
divider options (÷1 default condition, ÷2, ÷3, ÷4, ÷5) that are
set by loading the master clock divider field in Register B with
the appropriate code (see ). Once the internal device master
clock (DMCLK) has been set using the master clock divider, the
sample rate and serial clock settings are derived from DMCLK.
Table 12. DMCLK (Internal) Rate Divider Settings
MCD2
MCD1
MCD0
DMCLK Rate
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
MCLK
MCLK/2
MCLK/3
MCLK/4
MCLK/5
MCLK
MCLK
MCLK
SERIAL CLOCK RATE DIVIDER
The AD73322L features a programmable serial clock divider
that allows users to match the serial clock (SCLK) rate of the
data to that of the DSP engine or host processor. The maximum
SCLK rate available is DMCLK, and the other available rates are
DMCLK/2, DMCLK/4, and DMCLK/8. The slowest rate
(DMCLK/8) is the default SCLK rate. The serial clock divider is
programmable by setting bits CRB:2–3. Table 13 shows the
serial clock rate corresponding to the various bit settings.
The SPORT can work at four different serial clock (SCLK) rates
chosen from DMCLK, DMCLK/2, DMCLK/4, or DMCLK/8,
where DMCLK is the internal or device master clock resulting
from the external or pin master clock being divided by the
master clock divider.
SPORT REGISTER MAPS
Table 13. SCLK Rate Divider Settings
There are two register banks for each codec in the AD73322L,
the control register bank and the data register bank. The control
register bank consists of eight read/write registers, each eight
bits wide. Table 16 shows the control register map for the
AD73322L. The first two control registers, CRA and CRB, are
reserved for controlling the SPORT. They hold settings for
parameters such as serial clock rate, internal master clock rate,
sample rate and device count. As both codecs are internally
cascaded, registers CRA and CRB on each codec must be
programmed with the same setting to ensure correct operation
(this is shown in the programming examples).
SCD1
SCD0
SCLK Rate
DMCLK/8
DMCLK/4
DMCLK/2
DMCLK
0
0
1
1
0
1
0
1
SAMPLE RATE DIVIDER
The AD73322L features a programmable sample rate divider
that allows users flexibility in matching the codec’s ADC and
DAC sample rates (decimation/interpolation rates) to the needs
of the DSP software. The maximum sample rate available is
DMCLK/256, which offers the lowest conversion group delay,
while the other available rates are DMCLK/512, DMCLK/1024,
and DMCLK/2048. The slowest rate (DMCLK/2048) is the
default sample rate. The sample rate divider is programmable by
setting bits CRB:0-1. Table 14 shows the sample rate
The other five registers, CRC through CRH, are used to hold
control settings for the ADC, DAC, reference, power control,
and gain tap sections of the device. It is not necessary for the
contents of CRC through CRH on each codec be similar.
Control registers are written to on the negative edge of SCLK.
The data register bank consists of two, 16-bit registers that are
the DAC and ADC registers.
corresponding to the various bit settings.
Table 14. Sample Rate Divider Settings
MASTER CLOCK DIVIDER
DIR1
DIR0
SCLK Rate
The AD73322L features a programmable master clock divider
that allows the user to reduce an externally available master
clock, at pin MCLK, by a ratio of 1, 2, 3, 4, or 5 to produce an
internal master clock signal (DMCLK) that is used to calculate
the sampling and serial clock rates. The master clock divider is
programmable by setting CRB:4-6. Table 12 shows the division
ratio corresponding to the various bit settings. The default
divider ratio is divide-by-one.
0
0
1
1
0
1
0
1
DMCLK/2048
DMCLK/1024
DMCLK/512
DMCLK/256
Rev. A | Page 19 of 48
AD73322L
DAC ADVANCE REGISTER
The loading of the DAC is internally synchronized with the
unloading of the ADC data in each sampling interval. The
default DAC load event happens one SCLK cycle before the
SDOFS flag is raised by the ADC data being ready. However,
this DAC load position can be advanced before this time by
modifying the contents of the DAC advance field in Control
Register E (CRE:0–4). The field is five bits wide, allowing
31 increments of weight 1/(FS × 32), as shown in Table 15.
The sample rate, fS, depends on the setting of both the MCLK
divider and the sample rate divider, as shown in Table 12 and
Table 14. In certain circumstances this DAC update adjustment
can reduce the group delay when the ADC and DAC are used to
process data in series. For more information about how the
DAC advance register can be used, see the section Configuring
an AD73322L to Operate in Mixed Mode.
NOTE: The DAC advance register should not be changed while
the DAC section is powered up.
Table 15. DAC Timing Control
DA4
DA3
DA2
DA1
DA0
Time Advance
0 s
1/(FS × 32) s
2/(FS × 32) s
30/(FS × 32) s
31/(FS × 32) s
0
0
0
1
1
0
0
0
1
1
0
0
0
1
1
0
0
1
1
1
0
1
0
0
1
Table 16. Control Register Map
Address (Binary)
Name
Description
Type
Width
Reset Setting (Hex)
000
001
010
011
100
101
110
111
CRA
CRB
CRC
CRD
CRE
CRF
CRG
CRH
Control Register A
Control Register B
Control Register C
Control Register D
Control Register E
Control Register F
Control Register G
Control Register H
R/W
R/W
R/W
R/W
R/W
R/W
R/W
R/W
8
8
8
8
8
8
8
8
0x00
0x00
0x00
0x00
0x00
0x00
0x00
0x00
Table 17. Control Word Description
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
C/D
R/W
Device Address
Register Address
Register Data
Control
Frame
Description
Bit 15
Control/Data
When set high, this bit signifies a control word in program or mixed program/data modes. When set
low, it signifies a data-word in mixed program/data mode or an invalid control word in program mode.
Bit 14
Read/Write
When set low, this bit tells the device that the data field is to be written to the register selected by the
register field setting, provided the address field is zero. When set high, it tells the device that the
selected register is to be written to the data field in the input serial register and that the new control
word is to be output from the device via the serial output.
Bits 13 to 11
Device Address
This 3-bit field holds the address information. Only when this field is zero is a device selected. If the
address is not zero, it is decremented and the control word is passed out of the device via the serial
output.
Bits 10 to 8
Bits 7 to 0
Register Address
Register Data
This 3-bit field is used to select one of the eight control registers on the AD73322L.
This 8-bit field holds the data that is to be written to or read from the selected register provided the
address field is zero.
Rev. A | Page 20 of 48
AD73322L
CONTROL REGISTER A
Table 18. Control Register A Description
7
6
5
4
3
2
1
0
RESET
DC2
DC1
DC0
SLB
BLB
MM
DATA/PGM
Bit
0
Name
Description
DATA/PGM
MM
DLB
SLB
DC0
Operating Mode (0 = program; 1 = data mode)
Mixed Mode (0 = off; 1 = enabled)
Digital Loop-Back Mode (0 = off; 1 = enabled)
SPORT Loop-Back Mode (0 = off; 1 = enabled)
Device Count (Bit 0)
1
2
3
4
5
DC1
Device Count (Bit 1)
6
DC2
Device Count (Bit 2)
7
RESET
Software Reset (0 = off; 1 = initiates reset)
CONTROL REGISTER B
Table 19. Control Register B Description
7
6
5
4
3
2
1
0
CEE
MCD2
MCD1
MCD0
SCD1
SCD0
DIR1
DIR0
Bit
0
1
2
3
4
5
6
7
Name
Description
DIR0
DIR1
Decimation/Interpolation Rate (Bit 0)
Decimation/Interpolation Rate (Bit 1)
Serial Clock Divider (Bit 0)
Serial Clock Divider (Bit 1)
Master Clock Divider (Bit 0)
Master Clock Divider (Bit 1)
Master Clock Divider (Bit 2)
Control Echo Enable (0 = off; 1 = enabled)
SCD0
SCD1
MCD0
MCD1
MCD2
CEE
CONTROL REGISTER C
Table 20. Control Register C Description
7
6
5
4
3
2
1
0
—
RU
PUREF
PUDAC
PUADC
PUIA
PUAGT
PU
Bit
0
1
2
3
4
5
6
7
Name
PU
Description
Power-Up Device (0 = power-down; 1 = power on)
Analog Gain Tap Power (0 = power-down; 1 = power on)
Input Amplifier Power (0 = power-down; 1 = power on)
ADC Power (0 = power-down; 1 = power on)
DAC Power (0 = power-down; 1 = power on)
REF Power (0 = power-down; 1 = power on)
PUAGT
PUIA
PUADC
PUDAC
PUREF
RU
REFOUT Use (0 = disable REFOUT; 1 = enable REFOUT)
Reserved, must be programmed to 0
—
Rev. A | Page 21 of 48
AD73322L
CONTROL REGISTER D
Table 21. Control Register D Description
7
6
5
4
3
2
1
0
MUTE
OGS2
OGS1
OGS0
RMOD
IGS2
IGS1
IGS0
Bit
0
1
Name
IGS0
IGS1
Description
Input Gain Select (Bit 0)
Input Gain Select (Bit 1)
Input Gain Select (Bit 2)
2
IGS2
3
4
5
6
RMOD
OGS0
OGS1
OGS2
MUTE
Reset ADC Modulator (0 = off; 1 = reset enabled)
Output Gain Select (Bit 0)
Output Gain Select (Bit 1)
Output Gain Select (Bit 2)
Output Mute (0 = mute off; 1 = mute enabled)
7
CONTROL REGISTER E
Table 22. Control Register E Description
7
6
5
4
3
2
1
0
—
DGTE
IBYP
DA4
DA3
DA2
DA1
DA0
Bit
0
1
2
3
4
5
6
7
Name
Description
DA0
DA1
DA2
DA3
DA4
IBYP
DGTE
—
DAC Advance Setting (Bit 0)
DAC Advance Setting (Bit 1)
DAC Advance Setting (Bit 2)
DAC Advance Setting (Bit 3)
DAC Advance Setting (Bit 4)
Interpolator Bypass (0 = bypass disabled; 1 = bypass enabled)
Digital Gain Tap Enable (0 = disabled; 1 = enabled)
Reserved (program to 0)
CONTROL REGISTER F
Table 23. Control Register F Description
7
6
5
4
3
2
1
0
ALB/AGTM
INV
SEEN/AGTE
AGTC4
AGTC3
AGTC2
AGTC1
AGTC0
Bit
0
1
2
3
Name
Description
AGTC0
AGTC1
AGTC2
AGTC3
AGTC4
SEEN/
AGTE
INV
ALB/
AGTM
Analog Gain Tap Coefficient (Bit 0)
Analog Gain Tap Coefficient (Bit 1)
Analog Gain Tap Coefficient (Bit 2)
Analog Gain Tap Coefficient (Bit 3)
Analog Gain Tap Coefficient (Bit 4)
Single-Ended Enable (0 = disabled; 1 = enabled)
Analog Gain Tap Enable (0 = disabled; 1 = enabled)
Input Invert (0 = disabled; 1 = enabled)
Analog Loopback of Output to Input (0 = disabled; 1 = enabled)
Analog Gain Tap Mute (0 = off; 1 = muted)
4
5
6
7
Rev. A | Page 22 of 48
AD73322L
CONTROL REGISTER G
Table 24. Control Register G Description
7
6
5
4
3
2
1
0
DGTC7
DGTC6
DGTC5
DGTC4
DGTC3
DGTC2
DGTC1
DGTC0
Bit
0
1
2
3
4
5
6
7
Name
Description
DGTC0
DGTC1
DGTC2
DGTC3
DGTC4
DGTC5
DGTC6
DGTC7
Digital Gain Tap Coefficient (Bit 0)
Digital Gain Tap Coefficient (Bit 1)
Digital Gain Tap Coefficient (Bit 2)
Digital Gain Tap Coefficient (Bit 3)
Digital Gain Tap Coefficient (Bit 4)
Digital Gain Tap Coefficient (Bit 5)
Digital Gain Tap Coefficient (Bit 6)
Digital Gain Tap Coefficient (Bit 7)
CONTROL REGISTER H
Table 25. Control Register H Description
7
6
5
4
3
2
1
0
DGTC15
DGTC14
DGTC13
DGTC12
DGTC11
DGTC10
DGTC9
DGTC8
Bit
0
1
2
3
4
5
6
7
Name
Description
DGTC8
DGTC9
Digital Gain Tap Coefficient (Bit 8)
Digital Gain Tap Coefficient (Bit 9)
Digital Gain Tap Coefficient (Bit 10)
Digital Gain Tap Coefficient (Bit 11)
Digital Gain Tap Coefficient (Bit 12)
Digital Gain Tap Coefficient (Bit 13)
Digital Gain Tap Coefficient (Bit 14)
Digital Gain Tap Coefficient (Bit 15)
DGTC10
DGTC11
DGTC12
DGTC13
DGTC14
DGTC15
Rev. A | Page 23 of 48
AD73322L
OPERATION
RESETTING THE AD73322L
PROGRAM (CONTROL) MODE
The
pin resets all the control registers. All registers are
In program mode, CRA:0 = 0, the user writes to the control
registers to set up the device for desired operation—SPORT
operation, cascade length, power management, input/output
gain, etc. In this mode, the 16-bit information packet sent to the
device by the DSP engine is interpreted as a control word whose
format is shown in Table 17. In this mode, the user must
address the device to be programmed using the address field of
the control word. This field is read by the device and if it is zero
(000 bin), the device recognizes the word as being addressed to
it. If the address field is not zero, it is then decremented and the
control word is passed out of the device—either to the next
device in a cascade or back to the DSP engine.
RESET
reset to zero, indicating that the default SCLK rate (DMCLK/8)
and sample rate (DMCLK/2048) are at a minimum to ensure
that slow speed DSP engines can communicate effectively. As
well as resetting the control registers using the
device can be reset using the RESET bit (CRA:7) in Control
Register A. Both hardware and software resets require four
pin, the
RESET
DMCLK cycles. On reset, DATA/
(CRA:0) is set to 0
PGM
(default condition) thus enabling program mode. The reset
conditions ensure that the device must be programmed to the
correct settings after power-up or reset. Following a reset, the
SDOFS is asserted 2048 DMCLK cycles after
going high.
RESET
The data that is output following reset and during program
mode is random and contains no valid information until either
data or mixed mode is set.
This 3-bit address format allows the user to uniquely address
any one of up to eight devices in a cascade; please note that this
addressing scheme is valid only in sending control information
to the device —a different format is used to send DAC data to
the device(s). As the AD73322L is a dual codec, it features two
separate device addresses for programming purposes. If the
AD73322L is used in a standalone configuration connected to
a DSP, the two device addresses correspond to 0 and 1. If the
AD73322L is configured in a cascade of multiple, dual, or
single codecs (AD73322L or AD73311), its device addresses
correspond with its hardwired position in the cascade.
POWER MANAGEMENT
The individual functional blocks of the AD73322L can be
enabled separately by programming the Power Control Register
CRC. It allows certain sections to be powered down if not
required, which adds to the device’s flexibility in that the user
need not incur the penalty of having to provide power for a
certain section if it is not necessary to the design. The power
control registers provide individual control settings for the
major functional blocks on each codec unit and also a global
override that allows all sections to be powered up by setting the
bit. Using this method the user could, for example, individually
enable a certain section, such as the reference (CRC:5), and
disable all others. The global power-up (CRC:0) can be used to
enable all sections, but if power-down is required using the
global control, the reference is still enabled, in this case, because
its individual bit is set. Refer to Table 21 for details of the
settings of CRC.
Following reset, when the SE pin is enabled, the codec responds
by raising the SDOFS pin to indicate that an output sample
event has occurred. Control words can be written to the device
to coincide with the data being sent out of the SPORT, as shown
in Figure 20, or they can lag the output words by a time interval
that should not exceed the sample interval. After reset, output
frame sync pulses occur at a slower default sample rate, which is
DMCLK/2048, until Control Register B is programmed, after
which the SDOFS pulses are set according to the contents of
DIR0-1. This allows slow controller devices to establish
communication with the AD73322L. During program mode,
the data output by the device is random and should not be
interpreted as ADC data.
NOTE: As both codec units share a common reference, the
reference control bits (CRC:5-7) in each SPORT are wire-OR’ed
to allow either device to control the reference.
SE
OPERATING MODES
There are three main modes of operation available on the
AD73322L: program, data, and mixed program/data modes.
Two other operating modes are typically reserved as diag-
nostic modes: digital and SPORT loop-back. The device
configuration—register settings—can be changed only in
program and mixed program/data modes. In all modes,
transfers of information to or from the device occur in 16-bit
packets; therefore the DSP engine’s SPORT is programmed for
16-bit transfers.
SCLK
SDOFS
SAMPLE WORD (DEVICE 2) SAMPLE WORD (DEVICE 1)
SDO
SDIFS
SDI
CONTROL WORD
(DEVICE 2)
CONTROL WORD
(DEVICE 1)
Figure 20. Interface Signal Timing for Control Mode Operation
Rev. A | Page 24 of 48
AD73322L
DATA MODE
MIXED PROGRAM/DATA MODE
Once the device has been configured by programming the
correct settings to the various control registers, the device
may exit program mode and enter data mode. This is done
This mode allows the user to send control words to the device
along with the DAC data. This permits adaptive control of the
device where control of the input/output gains, etc., can be
affected by interleaving control words along with the normal
flow of DAC data. The standard data frame remains 16 bits, but
the MSB is used as a flag bit to indicate whether the remaining
15 bits of the frame represent DAC data or control information.
In the case of DAC data, the 15 bits are loaded with MSB
justification and LSB set to 0 to the DAC register. Mixed mode
is enabled by setting the MM bit (CRA:1) to 1 and the
by programming the DATA/
(CRA:0) bit to a 1 and
PGM
MM (CRA:1) to 0. Once the device is in data mode, the 16-bit
input data frame is interpreted as DAC data rather than a
control frame. This data is therefore loaded directly to the DAC
register. As Figure 20 shows, because the entire input data frame
contains DAC data in data mode, the device relies on counting
the number of input frame syncs received at the SDIFS pin.
When that number equals the device count stored in the device
count field of CRA, the device knows that the present data
frame being received is its own DAC update data. When the
device is in normal data mode (that is, mixed mode disabled), it
must receive a hardware reset to reprogram any of the control
register settings.
DATA/
bit (CRA:0) to 1. In the case where control setting
PGM
changes are required during normal operation, this mode
allows the ability to load both control and data information
with the slight inconvenience of formatting the data. Note that
the output samples from the ADC will also have the MSB set to
zero to indicate it is a data-word.
In a single AD73322L configuration, each 16-bit data frame
sent from the DSP to the device is interpreted as DAC data, but
it is necessary to send two DAC words per sample period in
order to ensure the DAC update. Also, as the device count
setting defaults to 1, it must be set to 2 (001b) to ensure correct
update of both DACs on the AD73322L.
The section Configuring an AD73322L to Operate in Mixed
Mode details the initialization and operation of an AD73322L
operating in mixed mode. Note that it is not essential to load
the control registers in Program Mode before setting mixed
mode active. It is also possible to initiate mixed mode by
programming CRA with the first control word and then
interleaving control words with DAC data.
The section DAC Timing Control Example details the initial-
ization and operation of an AD73322L in normal data mode.
DIGITAL LOOP-BACK MODE
SE
This mode can be used for diagnostic purposes, allowing the
user to feed the ADC samples from the ADC register directly to
the DAC register. This forms a loop-back of the analog input to
the analog output by reconstructing the encoded signal using
the decoder channel. The serial interface continues to work,
which allows the user to control gain settings, SCLK frequency,
sample rate, etc. Only when DLB is enabled with mixed mode
operation can the user disable the DLB—otherwise the device
must be reset.
SCLK
SDOFS
SDO
ADC SAMPLE WORD
(DEVICE 2)
ADC SAMPLE WORD
(DEVICE 1)
SDIFS
SDI
SPORT LOOP-BACK MODE
DAC DATA WORD
(DEVICE 2)
DAC DATA WORD
(DEVICE 1)
This mode allows the user to verify the DSP interfacing and
connection by writing words to the SPORT of the devices and
have them returned back unchanged after a delay of 16 SCLK
cycles. The frame sync and data-word that are sent to the device
are returned via the output port. Again, SLB mode can only be
disabled when used in conjunction with mixed mode, otherwise
the device must be reset.
Figure 21. Interface Signal Timing for Data Mode Operation
Rev. A | Page 25 of 48
AD73322L
ANALOG LOOP-BACK MODE
In analog loop-back mode, the differential DAC output is
connected, via a loop-back switch, to the ADC input, as shown
in Figure 22. This mode allows the ADC channel to check
functionality of the DAC channel as the reconstructed output
signal can be monitored using the ADC as a sampler. analog
loop-back is enabled by setting the ALB bit (CRF:7).
Note that analog loop-back can only be enabled if the Analog
Gain Tap is powered down (CRC:1 = 0).
ANALOG
LOOP-BACK
SELECT
SINGLE-
ENDED
ENABLE
INVERTING
OP AMPS
INVERT
VFBN1
VINN1
0/38dB
PGA
V
REF
VINP1
VFBP1
V
REF
GAIN
±1
ANALOG GAIN
TAP POWERED
DOWN
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
REFOUT
REFCAP
REFERENCE
AD73322L
Figure 22. Analog Loop-Back Connectivity
Rev. A | Page 26 of 48
AD73322L
INTERFACING
The AD73322L can be interfaced to most modern DSP engines
using conventional serial port connections and an extra enable
control line. Both serial input and output data use an accom-
panying frame synchronization signal that is active high one
clock cycle before the start of the 16-bit word or during the last
bit of the previous word if transmission is continuous. The
serial clock (SCLK) is an output from the codec and is used
to define the serial transfer rate to the DSP’s Tx and Rx ports.
Two primary configurations can be used: the first is shown in
Figure 22 where the DSP’s Tx data, Tx frame sync, Rx data, and
Rx frame sync are connected to the codec’s SDI, SDIFS, SDO,
and SDOFS, respectively. This configuration, referred to as
indirectly coupled or nonframe sync loop-back, has the effect of
decoupling the transmission of input data from the receipt of
output data. The delay between receipt of codec output data and
transmission of input data for the codec is determined by the
DSP’s software latency.
SDIFS
SDI
TFS
CODEC1
CODEC2
DT
SCLK
SCLK
SDO
ADSP-21xx
DSP
AD73322L
CODEC
DR
RFS
SDOFS
Figure 23. Indirectly Coupled or Nonframe Sync
Loop-Back Configuration
CASCADE OPERATION
The AD73322L has been designed to support cascading of
codecs from a single DSP serial port (see Figure 36). Cascaded
operation can support mixes of dual- or single-channel devices
with the maximum number of codec units being eight (the
AD73322L is equivalent to two codec units). The SPORT
interface protocol has been designed so that device addressing
is built into the packet of information sent to the device. This
allows the cascade to be formed with no extra hardware
overhead for control signals or addressing. A cascade can be
formed in either of the two modes previously discussed.
When programming the DSP serial port for this configuration,
it is necessary to set the Rx FS as an input and the Tx FS as an
output generated by the DSP. This configuration is most useful
when operating in mixed mode, as the DSP has the ability to
decide how many words (either DAC or control) can be sent to
the codecs. This means that full control can be implemented
over the device configuration as well as updating the DAC in a
given sample interval.
There may be some restrictions in cascade operation due to the
number of devices configured in the cascade and the sampling
rate and serial clock rate chosen. The following relationship
details the restrictions in configuring a codec cascade.
The second configuration (shown in Figure 24) has the DSP’s
Tx data and Rx data connected to the codec’s SDI and SDO,
respectively, while the DSP’s Tx and Rx frame syncs are
connected to the codec’s SDIFS and SDOFS. In this
Number of Codes × Word Size (16) × Sampling Rate ≤
Serial Clock Rate
configuration, referred to as directly coupled or frame sync
loop-back, the frame sync signals are connected together and
the input data to the codec is forced to be synchronous with the
output data from the codec. The DSP must be programmed so
that both the Tx FS and Rx FS are inputs as the codec SDOFS is
input to both. This configuration guarantees that input and
output events occur simultaneously and is the simplest
configuration for operation in normal data mode. When
programming the DSP in this configuration, it is advisable to
preload the Tx register with the first control word to be sent
before the codec is taken out of reset. This ensures that this
word is transmitted to coincide with the first output word from
the device(s).
TFS
SDIFS
SDI
CODEC1
CODEC2
DT
SCLK
SCLK
SDO
ADSP-21xx
DSP
AD73322L
CODEC
DR
RFS
SDOFS
Figure 24. Directly Coupled or Frame Sync Loop-Back Configuration
When using the indirectly coupled frame sync configuration
in cascaded operation, be aware of the restrictions in sending
data to all devices in the cascade. Effectively the time allowed is
given by the sampling interval (M/DMCLK—where M can be
256, 512, 1024, or 2048), which is 125 µs for a sample rate of
8 kHz. In this interval, the DSP must transfer N × 16 bits of
information where N is the number of devices in the cascade.
Rev. A | Page 27 of 48
AD73322L
Each bit will take 1/SCLK and, allowing for any latency between
the receipt of the Rx interrupt and the transmission of the Tx
data, the relationship for successful operation is given by
Table 26. Device Count Settings
DC2
DC1
DC0
Cascade Length
0
0
0
0
0
1
1
2
3
4
5
6
7
8
M/DMCLK > ((N × 16/SCLK) + TINTERRUPT LATENCY
)
0
0
The interrupt latency will include the time between the ADC
sampling event and the Rx interrupt being generated in the
DSP—this should be 16 SCLK cycles.
0
1
1
1
1
0
0
1
1
1
0
1
0
Because the AD73322L is configured in cascade mode, each
device must know the number of devices in the cascade because
the data and mixed modes use a method of counting input
frame sync pulses to decide when they should update the DAC
register from the serial input register. Control Register A
contains a 3-bit field (DC0-2) that is programmed by the DSP
during the programming phase. The default condition is that
the field contains 000b, which is equivalent to a single device in
the cascade (see Table 26). However, for cascade operation this
field must contain a binary value that is one less than the
number of devices in the cascade, which is 001b for a single
AD73322L device configuration.
1
1
Rev. A | Page 28 of 48
AD73322L
PERFORMANCE
Because the AD73322L is designed to provide high perfor-
mance and low cost conversion, it is important to understand
how high performance can be achieved in a typical application.
By means of spectral graphs, this section outlines the typical
performance of the device and highlights some of the options
available to users in achieving their desired sample rate, either
directly in the device or by doing some post-processing in the
DSP, while also showing the advantages and disadvantages of
the different approaches.
The decimator’s frequency response (Sinc3) gives some pass-
band attenuation (up to FS/2) which continues to roll off above
the Nyquist frequency. If it is required to implement a digital
filter to create a sharper cutoff characteristic, it may be prudent
to use an initial sample rate of greater than twice the Nyquist
rate in order to avoid aliasing due to the smooth roll-off of the
sinc3 filter response.
0
–20
–40
ENCODER SECTION
The AD73322L offers a variable sampling rate from a fixed
MCLK frequency—with 64 kHz, 32 kHz, 16 kHz, and 8 kHz
being available with a 16.384 MHz external clock. Each of
these sampling rates preserves the same sampling rate in the
ADC’s sigma-delta modulator, which ensures that the noise
performance is optimized in each case. The examples that
follow show the performance of a 1 kHz sine wave when
converted at the various sample rates.
–60
–80
–100
–120
–140
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4
FREQUENCY (Hz)
×10
The range of sampling rates is aimed to offer the user a degree
of flexibility in deciding how the analog front end is to be
implemented. The high sample rates of 64 kHz and 32 kHz are
suited to those applications, such as active control, where low
conversion group delay is essential. On the other hand, the
lower sample rates of 16 kHz and 8 kHz are better suited for
applications such as telephony, where the lower sample rates
result in lower DSP overhead.
Figure 25. FFT (ADC 64 kHz Sampling)
0
–20
–40
–60
Figure 29 shows the spectrum of the 1 kHz test tone sampled
at 64 kHz. The plot shows the characteristic shaped noise floor
of a sigma-delta converter, which is initially flat in the band of
interest but then rises with increasing frequency. If a suitable
digital filter is applied to this spectrum, the noise floor can be
eliminated in the higher frequencies. This signal can then be
used in DSP algorithms or can be further processed in a
decimation algorithm to reduce the effective sample rate.
Figure 26 shows the resulting spectrum following the filtering
and decimation of the spectrum of Figure 25 from 64 kHz to
an 8 kHz rate.
–80
–100
–120
0
500
1000
1500
2000
2500
3000
3500
4000
FREQUENCY (Hz)
Figure 26. FFT (ADC 8 kHz Filtered and Decimated from 64 kHz)
0
50
100
150
The AD73322L also features direct sampling at the lower rate of
8 kHz. This is achieved by the use of extended decimation
registers within the decimator block, which allows for the
increased word growth associated with the higher effective
oversampling ratio. Figure 27 details the spectrum of a 1 kHz
test tone converted at an 8 kHz rate.
The device features an on-chip, master clock divider circuit that
allows the sample rate to be reduced because the sampling rate
of the sigma-delta converter is proportional to the output of the
MCLK Divider (whose default state is divide-by-one).
0
500
1000
1500
2000
2500
3000
3500
4000
FREQUENCY (Hz)
Figure 27. FFT (ADC 8 kHz Direct Sampling)
Rev. A | Page 29 of 48
AD73322L
Consider a second example:
Group Delay (decimator @ 64 kHz) = 3 × (32 − 1)/2 ×
(1/2.048e6) = 22.7 µs
In the case of voice-band processing where 4 kHz represents the
Nyquist frequency, if the signal to be measured were externally
band-limited, then an 8 kHz sampling rate would suffice.
However, if the bandwidth must be limited with a digital filter,
then it may be more appropriate to use an initial sampling rate
of 16 kHz and to process this sample stream with a filtering and
decimating algorithm to achieve a 4 kHz band-limited signal at
an 8 kHz rate. Figure 19 details the initial 16 kHz sampled tone.
If final filtering is implemented in the DSP, the final filter’s
group delay must be taken into account when calculating
overall group delay.
DECODER SECTION
The decoder section updates (samples) at the same rate as the
encoder section. This rate is programmable as 64 kHz, 32 kHz,
16 kHz, or 8 kHz (from a 16.384 MHz MCLK). The decoder
section represents a reverse of the process that was described in
the encoder section. In the case of the decoder section, signals
are applied in the form of samples at an initial low rate. This
sample rate is then increased to the final digital sigma-delta
modulator rate of DMCLK/8 by interpolating new samples
between the original samples. The interpolating filter also has
the action of canceling images due to the interpolation process
using spectral nulls that exist at integer multiples of the initial
sampling rate. Figure 30 shows the spectral response of the
decoder section sampling at 64 kHz. Again, its sigma-delta
modulator shapes the noise so it is reduced in the voice
bandwidth dc–4 kHz. For improved voice-band SNR, the user
can implement an initial anti-imaging filter, preceded by 8 kHz
to 64 kHz interpolation, in the DSP.
0
–20
–40
–60
–80
–100
–120
–140
0
1000
2000
3000
4000
5000
6000
7000
8000
FREQUENCY (Hz)
Figure 28. FFT (ADC 16 kHz Direct Sampling)
Figure 29 shows the spectrum of the final 8 kHz sampled
filtered tone.
0
–10
–20
–30
–40
–50
–60
–70
–80
–90
–100
0
–20
–40
–60
–80
–100
–120
–140
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4
FREQUENCY (Hz)
×10
0
500
1000
1500
2000
2500
3000
3500
4000
FREQUENCY (Hz)
Figure 30. FFT (DAC 64 kHz Sampling)
Figure 29. FFT (ADC 8 kHz Filtered and Decimated from 16 kHz)
ENCODER GROUP DELAY
When programmed for high sampling rates, the AD73322L
offers a very low level of group delay, which is given by
Group Delay (Decimator) = Order × ((M − 1)/2) × TDEC
where:
Order is the order of the decimator (= 3)
M is the decimation factor (= 32 @ 64 kHz, = 64 @ 32 kHz,
= 128 @ 16 kHz , = 256 @ 8 kHz)
T
DEC is the decimation sample interval (= 1/2.048e6 based on
DMCLK = 16.384 MHz)
Rev. A | Page 30 of 48
AD73322L
Because the AD73322L can be operated at 8 kHz (see Figure 31)
or 16 kHz sampling rates, which make it particularly suited for
voice-band processing, the user must understand the action of
the interpolator’s sinc3 response. As was the case with the
encoder section, if the output signal’s frequency response is not
bounded by the Nyquist frequency, it may be necessary to
perform some initial digital filtering to eliminate signal energy
above Nyquist to ensure that it is not imaged at the integer
multiples of the sampling frequency. If the user chooses to
bypass the interpolator, perhaps to reduce group delay, images
of the original signal are generated at integer intervals of the
sampling frequency. In this case these images must be removed
by external analog filtering.
DMCLK of 16.384 MHz), and the simple, external RC antialias
filter is sufficient to provide the required stop-band rejection
above the Nyquist frequency for this sample rate. In the case of
the ADC section, the decimating filter is required to both
decrease sample rate and increase sample resolution. The
process of changing sample rate (resampling) leads to aliases of
the original sampled waveform appearing at integer multiples of
the new sample rate. These aliases would get mapped into the
required signal pass band without the application of some
further antialias filtering. In the AD73322L, the sinc-cubed
response of the decimating filter creates spectral nulls at integer
multiples of the new sample rate. These nulls coincide with the
aliases of the original waveform, which were created by the
down-sampling process, therefore reducing or eliminating the
aliasing due to sample rate reduction.
0
–10
–20
–30
–40
–50
–60
–70
–80
–90
–100
In the DAC section, increasing the sampling rate by
interpolation creates images of the original waveform at
intervals of the original sampling frequency. These images may
be sufficiently rejected by external circuitry but the sinc-cubed
filter in the interpolator again nulls the output spectrum at
integer intervals of the original sampling rate, which
corresponds with the images due to the interpolation process.
The spectral response of a sinc-cubed filter shows the charac-
teristic nulls at integer intervals of the sampling frequency. Its
pass-band characteristic (up to Nyquist frequency) features a
roll-off that continues up to the sampling frequency, where the
first null occurs. In many applications this smooth response
does not give sufficient attenuation of frequencies outside the
band of interest; therefore, it may be necessary to implement a
final filter in the DSP to equalize the pass-band roll-off and
provide a sharper transition band and greater stop-band
attenuation.
0
500
1000
1500
2000
2500
3000
3500
4000
FREQUENCY (Hz)
Figure 31. FFT (DAC 8 kHz Sampling)
Figure 32 shows the output spectrum of a 1 kHz tone generated
at an 8 kHz sampling rate with the interpolator bypassed.
0
–10
–20
–30
–40
–50
–60
–70
–80
–90
–100
DECODER GROUP DELAY
The interpolator roll-off is mainly due to its sinc-cubed
function characteristic, which has an inherent group delay given
by the equation
Group Delay (Interpolator) = Order × (L − 1)/2) × TINT
where:
Order is the interpolator order (= 3).
L is the interpolation factor (= 32 @ 64 kHz, = 64 @ 32 kHz,
= 128 @ 16 kHz, = 256 @ 8 kHz).
0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4
FREQUENCY (Hz)
×10
TINT is the interpolation sample interval (= 1/2.048e6).
Figure 32. FFT (DAC 8 kHz Sampling—Interpolator Bypassed)
Consider a second example:
Group Delay (Interpolator @ 64 kHz) = 3 × (32 − 1)/2 ×
(1/2.048e6) = 22.7 µs
ON-CHIP FILTERING
The analog section has a group delay of approximately 25 µs.
The primary function of the system filtering’s sinc-cubed
(Sinc3) response is to eliminate aliases or images of the ADCs
or DAC’s resampling, respectively. Both modulators are sampled
at a nominal rate of DMCLK/8 (which is 2.048 MHz for a
Rev. A | Page 31 of 48
AD73322L
DESIGN CONSIDERATIONS
The AD73322L features both differential inputs and outputs
on each channel to provide optimal performance and avoid
common-mode noise. It is also possible to interface either
inputs or outputs in single-ended mode. This section details the
choice of input and output configurations and also gives some
tips towards successful configuration of the analog interface
sections.
The primary concerns in interfacing to the ADC are, first, to
provide adequate antialias filtering and to ensure that the signal
source drives the switched-capacitor input of the ADC
correctly. The sigma-delta design of the ADC and its over-
sampling characteristics simplify the antialias requirements, but
the single-pole RC filter is primarily intended to eliminate
aliasing of frequencies above the Nyquist frequency of the
sigma-delta modulator’s sampling rate (typically 2.048 MHz). It
may still require a more specific digital filter implementation in
the DSP to provide the final signal-frequency response
characteristics.
ANTI-ALIAS
FILTER
VFBN1
100Ω
VINN1
0.047µF
0/38dB
V
PGA
REF
For optimum performance, the capacitors used for the
antialiasing filter must be of high quality dielectric (NPO). A
second concern is interfacing the signal source to the ADC’s
switched capacitor input load. The SC input presents a complex
dynamic load to a signal source, therefore, note that the slew
rate characteristic is an important consideration when choosing
external buffers for use with the AD73322L. The internal
inverting op amps on the AD73322L are specifically designed
to interface to the ADC’s SC input stage.
0.047µF
100Ω
VINP1
V
REF
VFBP1
GAIN
±1
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
REFOUT
REFCAP
REFERENCE
The AD73322L’s on-chip 38 dB preamplifier can be enabled
when there is not enough gain in the input circuit; the pre-
amplifier is configured by bits IGS0-2 of CRD. The total gain
must be configured to ensure that a full-scale input signal
produces a signal level at the input to the sigma-delta
modulator of the ADC that does not exceed the maximum
input range.
0.1µF
AD73322L
Figure 33. Analog Input (DC-Coupled)
ANALOG INPUTS
There are several different ways in which the analog input
(encoder) section of the AD73322L can be interfaced to
external circuitry. It provides optional input amplifiers which
allow sources with high source impedance to drive the ADC
section correctly. When the input amplifiers are enabled, the
input channel is configured as a differential pair of inverting
amplifiers referenced to the internal reference (REFCAP) level.
The inverting terminals of the input amplifier pair are
designated as Pins VINP1 and VINN1 for Channel 1 (VINP2
and VINN2 for Channel 2). The amplifier feedback connections
are available on Pins VFBP1 and VFBN1 for Channel 1 (VFBP2
and VFBN2 for Channel 2).
The dc biasing of the analog input signal is accomplished with
an on-chip voltage reference. If the input signal is not biased
at the internal reference level (via REFOUT), then it must be
ac-coupled with external coupling capacitors. CIN should be
0.1 µF or larger. The dc biasing of the input can then be
accomplished using resistors to REFOUT, as Figure 36 and
Figure 37 show.
ANTI-ALIAS
FILTER
VFBN1
100Ω
VINN1
0.047µF
0/38dB
V
PGA
REF
For applications where external signal buffering is required,
the input amplifiers can be bypassed and the ADC driven
directly. When the input amplifiers are disabled, the sigma-
delta modulator’s input section (SC PGA) is accessed directly
through the VFBP1 and VFBN1 pins for Channel 1 (VFBP2
and VFBN2 for Channel 2).
0.047µF
100Ω
VINP1
V
REF
VFBP1
GAIN
±1
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
OPTIONAL
BUFFER
It is also possible to drive the ADCs in either differential or
single-ended modes. If the single-ended mode is chosen, it is
possible using software control to multiplex between two single-
ended inputs connected to the positive and negative input pins.
REFOUT
REFCAP
REFERENCE
0.1µF
AD73322L
Figure 34. Analog Input (DC-Coupled) Using External Amplifiers
Rev. A | Page 32 of 48
AD73322L
The AD73322L’s ADC inputs are biased about the internal
reference level (REFCAP level); therefore, it may be necessary
to bias external signals to this level using the buffered REFOUT
level as the reference. This is applicable in either dc-coupled or
ac-coupled configurations. In the case of dc coupling, the signal
(biased to REFOUT) may be applied directly to the inputs
(using amplifier bypass), as shown in Figure 33, or it may be
conditioned in an external op amp where it can also be biased
to the reference level using the buffered REFOUT signal, as
shown in Figure 34, or it is possible to connect inputs directly to
the AD73322L’s input op amps as shown in Figure 35.
If the ADC is being connected in single-ended mode, the
AD73322L should be programmed for single-ended mode
using the SEEN and INV bits of CRF and the inputs connected
as shown in Figure 37. When operated in single-ended input
mode, the AD73322L can multiplex one of the two inputs to the
ADC input.
0.1µF
100Ω
VFBN1
VINN1
0.047µF
10kΩ
0/38dB
V
PGA
REF
VINP1
V
REF
VFBP1
100pF
GAIN
±1
50kΩ
VFBN1
VINN1
50kΩ
0/38dB
PGA
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
V
REF
+6/–15dB
PGA
50kΩ
VINP1
V
REF
VFBP1
50kΩ
REFOUT
REFCAP
REFERENCE
GAIN
±1
100pF
0.1µF
AD73322L
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
Figure 37. Analog Input (AC-Coupled) Single-Ended
+6/–15dB
PGA
If best performance is required from a single-ended source, it is
possible to configure the AD73322L’s input amplifiers as a
single-ended-to-differential converter, as shown in Figure 38.
REFOUT
REFCAP
REFERENCE
0.1µF
AD73322L
100pF
50kΩ
Figure 35. Analog Input (DC Coupled) Using Internal Amplifiers
VFBN1
VINN1
50kΩ
In the case of ac coupling, a capacitor is used to couple the
signal to the input of the ADC. The ADC input must be biased
to the internal reference (REFCAP) level which is done by
connecting the input to the REFOUT pin through a 10 kΩ
resistor, as shown in Figure 36.
0/38dB
V
PGA
REF
50kΩ
VINP1
V
REF
VFBP1
50kΩ
GAIN
±1
100pF
0.1µF
100Ω
VFBN1
VINN1
0.047µF
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
0/38dB
PGA
10kΩ
V
REF
VINP1
0.1µF
10kΩ
100Ω
REFOUT
REFCAP
V
REFERENCE
REF
VFBP1
0.047µF
GAIN
±1
0.1µF
AD73322L
Figure 38. Single-Ended-to-Differential Conversion on Analog Input
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
REFOUT
REFCAP
REFERENCE
0.1µF
AD73322L
Figure 36. Analog Input (AC-Coupled) Differential
Rev. A | Page 33 of 48
AD73322L
INTERFACING TO AN ELECTRET MICROPHONE
VFBN1
VINN1
Figure 39 details an interface for an electret microphone which
may be used in some voice applications. Electret microphones
typically feature a FET amplifier whose output is accessed on
the same lead which supplies power to the microphone;
therefore, this output signal must be capacitively coupled to
remove the power supply (dc) component. In this circuit, the
AD73322L input channel is being used in single-ended mode
where the internal inverting amplifier provides suitable gain to
scale the input signal relative to the ADC’s full-scale input
range. The buffered internal reference level at REFOUT is used
via an external buffer to provide power to the electret
microphone. This provides a quiet, stable supply for the
microphone. If this is not a concern, then the microphone can
be powered from the system power supply.
V
REF
VINP1
VFBP1
GAIN
±1
C
C
OUT
OUT
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
R
LOAD
REFOUT
REFCAP
REFERENCE
C
REFCAP
AD73322L
5V
Figure 40. Example Circuit for Differential Output
R
A
10µF
100pF
Figure 41 shows an example circuit for providing a single-ended
output with ac coupling. The capacitor of this circuit (COUT) is
not optional if dc current drain is to be avoided.
50kΩ
VFBN1
VINN1
C2
R
R1
B
ELECTRICITY
PROBE
0/38dB
PGA
V
REF
VINP1
VFBN1
VINN1
V
REF
VFBP1
GAIN
±1
V
REF
VINP1
VFBP1
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
+6/–15dB
PGA
GAIN
±1
REFOUT
REFCAP
REFERENCE
C
OUT
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
C
+6/–15dB
PGA
R
REFCAP
LOAD
AD73322L
REFOUT
REFCAP
Figure 39. Electret Microphone Interface Circuit
REFERENCE
ANALOG OUTPUT
0.1µF
AD73322L
The AD73322L’s differential analog output (VOUT) is produced
by an on-chip differential amplifier. The differential output can
be ac-coupled or dc-coupled directly to a load which can be a
headset or the input of an external amplifier (the specified
minimum resistive load on the output section is 150 Ω.) It is
possible to connect the outputs in either a differential or a
single-ended configuration, but please note that the effective
maximum output voltage swing (peak to peak) is halved in the
case of single-ended connection. Figure 40 shows a simple
circuit providing a differential output with ac coupling. The
capacitors in this circuit (COUT) are optional; if used, their value
can be chosen as follows:
Figure 41. Example Circuit for Single-Ended Output
1
COUT
=
2π fc RLOAD
where fC = desired cutoff frequency.
Rev. A | Page 34 of 48
AD73322L
FSX
SDIFS
SDI
DIFFERENTIAL-TO-SINGLE-ENDED OUTPUT
DT
In some applications it may be desirable to convert the full
differential output of the decoder channel to a single-ended
signal. The circuit of Figure 42 shows a scheme for doing this.
CLKX
SCLK
CLKR
TMS320C5x
DSP
AD73322L
CODEC
DR
SDO
SDOFS
RESET
SE
FSR
XF
0/38dB
V
PGA
REF
VINP1
V
REF
VFBP1
GAIN
±1
Figure 44. AD73322L Connected to TMS320C5x
R
F
VOUTP1
VOUTN1
CONTINUOUS
TIME
LOW-PASS
FILTER
CASCADE OPERATION
R1
R1
+6/–15dB
PGA
R
R
F
LOAD
Where it is required to configure a cascade of up to eight codecs
(four AD73322L dual codecs), ensure that the timing of the SE
REFOUT
REFCAP
REFERENCE
and
signals is synchronized at each device in the
RESET
0.1µF
AD73322L
cascade. A simple D-type flip-flop is sufficient to sync each
signal to the master clock MCLK, as in Figure 45.
Figure 42. Example Circuit for Differential to Single-Ended
Output Conversion
DSP CONTROL
TO SE
SE SIGNAL SYNCHRONIZED
TO MCLK
D
Q
1/2
74HC74
DIGITAL INTERFACING
MCLK
The AD73322L is designed to interface easily to most common
DSPs. The SCLK, SDO, SDOFS, SDI, and SDIFS must be con-
nected to the DSP’s serial clock, receive data, receive data
frame sync, transmit data, and transmit data frame sync pins,
respectively. The SE pin may be controlled from a parallel
output pin or flag pin such as FL0-2 on the ADSP-21xx (or XF
on the TMS320C5x) or, where SPORT power-down is not
required, it can be permanently strapped high using a suitable
CLK
DSP CONTROL
TO RESET
RESET SIGNAL SYNCHRONIZED
TO MCLK
D
Q
1/2
74HC74
MCLK
CLK
RESET
Figure 45. SE and
Sync Circuit or Cascaded Operation
pull-up resistor. The
pin may be connected to the
RESET
system hardware reset structure or it may also be controlled
using a dedicated control line. In the event of tying it to the
global system reset, it is advisable to operate the device in mixed
mode, which allows a software reset, otherwise there is no
convenient way of resetting the device. Figure 43 and Figure 44
show typical connections to an ADSP-218x and TMS320C5x,
respectively.
Connection of a cascade of devices to a DSP, as shown in
Figure 46, is no more complicated than connecting a single
device. Instead of connecting the SDO and SDOFS to the DSP’s
Rx port, these are now daisy-chained to the SDI and SDIFS of
the next device in the cascade. The SDO and SDOFS of the final
device in the cascade are connected to the DSP’s Rx port to
complete the cascade. SE and
on all devices are fed from
RESET
the signals that were synchronized with the MCLK using the
circuit, as described previously. The SCLK from only one device
need be connected to the DSP’s SCLK input(s) as all devices run
at the same SCLK frequency and phase.
TFS
DT
SDIFS
SDI
SCLK
SCLK
ADSP-218x
DSP
AD73322L
CODEC
DR
SDO
RFS
SDOFS
RESET
SE
FL0
FL1
Figure 43. AD73322L Connected to ADSP-218x
Rev. A | Page 35 of 48
AD73322L
GROUNDING AND LAYOUT
DIGITAL GROUND
Because the analog inputs to the AD73322L are differential,
most of the voltages in the analog modulator are common-
mode voltages. The excellent common-mode rejection of the
part removes common-mode noise on these inputs. The analog
and digital supplies of the AD73322L are independent and
separately pinned out to minimize coupling between analog and
digital sections of the device. The digital filters on the encoder
section will provide rejection of broadband noise on the power
supplies, except at integer multiples of the modulator sampling
frequency. The digital filters also remove noise from the analog
inputs provided the noise source does not saturate the analog
modulator. However, because the resolution of the AD73322L’s
ADC is high, and the noise levels from the AD73322L are so
low, care must be taken with regard to grounding and layout.
ANALOG GROUND
Figure 47. Ground Plane Layout
Avoid running digital lines under the device because they
couple noise onto the die. The analog ground plane should be
allowed to run under the AD73322L to avoid noise coupling.
The power supply lines to the AD73322L should use as large a
trace as possible to provide low impedance paths and reduce the
effects of glitches on the power supply lines. Fast switching
signals, such as clocks, should be shielded with digital ground
to avoid radiating noise to other sections of the board. Clock
signals should never be run near the analog inputs. Traces on
opposite sides of the board should run at right angles to each
other. This reduces the effects of feedthrough on the board. A
microstrip technique is by far the best to use, but is not always
possible with a double-sided board. In this technique, the
component side of the board is dedicated to ground planes,
while signals are placed on the other side.
TFS
SDIFS
SDI
MCLK
SE
DT
ADSP-218x
DSP
SCLK
AD73322L
RESET
SCLK
SDO
CODEC
DR
RFS
SDOFS
DEVICE 1
FL0
FL1
Good decoupling is important when using high speed devices.
On the AD73322L, both the reference (REFCAP) and supplies
need to be decoupled. It is recommended that the decoupling
capacitors used on both REFCAP and the supplies be placed as
close as possible to their respective pins to ensure high
performance from the device. All analog and digital supplies
should be decoupled to AGND and DGND respectively, with
0.1 µF ceramic capacitors in parallel with 10 µF tantalum
capacitors. In systems where a common-supply voltage is used
to drive both the AVDD and DVDD of the AD73322L, it is
recommended that the system’s AVDD supply be used. This
supply should have the recommended analog supply decoupling
between the AVDD pins of the AD73322L and AGND and the
recommended digital supply decoupling capacitors between the
DVDD pin and DGND.
MCLK
SE
SDIFS
SDI
AD73322L RESET
CODEC
SCLK
SDO
SDOFS
DEVICE 2
D1
D2
Q1
Q2
74HC74
Figure 46. Connection of Two AD73322Ls Cascaded to ADSP-218x
The printed circuit board that houses the AD73322L should be
designed so the analog and digital sections are separated and
confined to certain sections of the board. The AD73322L pin
configuration offers a major advantage in that its analog and
digital interfaces are connected on opposite sides of the
package. This facilitates the use of ground planes that can be
easily separated, as shown in Figure 47. A minimum etch
technique is generally best for ground planes because it gives
the best shielding. Digital and analog ground planes should be
joined in only one place. If this connection is close to the
device, it is recommended a ferrite bead inductor be used, as
shown in Figure 47.
Rev. A | Page 36 of 48
AD73322L
DSP PROGRAMMING CONSIDERATIONS
Autobuffering allows the user to specify the number of input or
output words (samples) that are transferred before a specific Tx
or Rx SPORT interrupt is generated. Given that the AD73322L
outputs two sample words per sample period, it is possible,
using auto-buffering, to have the DSP’s SPORT generate a single
interrupt on receipt of the second of the two sample words.
Additionally, both samples could be stored in a data buffer
within the data memory store. This technique has the advantage
of reducing the number of both Tx and Rx SPORT interrupts to
a single one at each sample interval. The user also knows where
each sample is stored. The alternative is to handle a larger
number of SPORT interrupts (twice as many in the case of a
single AD73322L) while also having some status flags to
indicate the origin and destination of each new sample.
This section discusses how the serial port of the DSP should
be configured and the implications of whether Rx and Tx
interrupts should be enabled.
DSP SPORT CONFIGURATION
Following are the key settings of the DSP SPORT required for
the successful operation with the AD73322L:
•
•
•
•
•
Configure for external SCLK
Serial word length = 16 bits
Transmit and receive frame syncs required with every word
Receive frame sync is an input to the DSP
Transmit frame sync is an:
Input—in frame sync loop-back mode
Output—in nonframe sync loop-back mode
MIXED-MODE OPERATION
To take full advantage of mixed-mode operation, configure the
DSP/Codec interface in nonFSLB and disable autobuffering.
This allows a variable number of words to be sent to the
AD73322L in each sample period—the extra words being
control words that are typically used to update gain settings in
adaptive control applications. The recommended sequence for
updating control registers in mixed mode is to send the control
word(s) first before the DAC update word.
•
•
Frame syncs occur one SCLK cycle before the MSB of the
serial word
Frame syncs are active high
DSP SPORT INTERRUPTS
If SPORT interrupts are enabled, it is important to note that
the active signals on the frame sync pins do not necessarily
correspond in real time to when SPORT interrupts are
generated.
It is possible to use mixed-mode operation when configured
in FSLB, but it is necessary to replace the DAC update with a
control word write in each sample period. This may cause some
discontinuity in the output signal due to a sample point being
missed and the previous sample being repeated. However, this
may be acceptable in some cases as the effect may be masked by
gain changes, etc.
On ADSP-21xx processors, it is necessary to enable SPORT
interrupts and use interrupt service routines (ISRs) to handle
Tx/Rx activity, while on the TMS320CSx processors, it is
possible to poll the status of the Rx and Tx registers. This means
that Rx/Tx activity can be monitored using a single ISR that
would ideally be the Tx ISR because the Tx interrupt typically
occurs before the Rx ISR.
INTERRUPTS
The AD73322L transfers and receives information over the
serial connection from the DSP’s SPORT. This occurs following
reset—during the initialization phase—and in both data mode
and mixed mode. Each transfer of data to or from the DSP can
cause a SPORT interrupt to occur. However even in FSLB
configuration where serial transfers in and out of the DSP are
synchronous, Tx and Rx interrupts do not occur at the same
time due to the way that Tx and Rx interrupts are generated
internally within the DSP’s SPORT. This is especially important
in time-critical, control loop applications where it may be
necessary to use Rx interrupts only, as the relative positioning of
the Tx interrupts relative to the Rx interrupts in a single sample
interval are not suitable for quick update of new DAC positions.
DSP SOFTWARE CONSIDERATIONS WHEN
INTERFACING TO THE AD73322L
When choosing the operating mode and hardware config-
uration of the AD73322L, be aware of their implications for
DSP software operation. The user has the flexibility of choosing
from either FSLB or nonFSLB when deciding on DSP-to-AFE
connectivity. There is also a choice to be made between using
autobuffering of input and output samples, or simply choosing
to accept them as individual interrupts. Because most modern
DSP engines support these modes, this section discusses these
topics in a generic DSP sense.
OPERATING MODE
The AD73322L supports two basic operating modes: frame
sync loop back (fslb) and nonfslb (see the Interfacing section).
As described previously, FSLB has some limitations when used
in mixed mode but is very suitable for use with the
autobuffering feature that is offered on many modern DSPs.
Rev. A | Page 37 of 48
AD73322L
INITIALIZATION
Following reset, the AD73322L is in its default condition, which
ensures that the device is in control mode and must be
programmed or initialized from the DSP to start conversions.
Because communications between AD73322L and the DSP are
interrupt driven, it is usually not practical to embed the initial-
ization codes into the body of the initialization routine. It is
more practical to put the sequence of initialization codes in a
data (or program) memory buffer and to access this buffer with
a pointer that is updated on each interrupt. If a circular buffer is
used, it allows the interrupt routine to check when the circular
buffer pointer has wrapped around—at which point the
initialization sequence is complete.
Hard-coding involves creating a sequence of writes to the DSP’s
SPORT Tx buffer, which are separated by loops or instructions
that idle and wait for the next Tx interrupt to occur, as shown in
the code that follows.
ax0 = b#1000100100000100;
tx0 = ax0;
idle; {wait for tx register to send current word}
The circular buffer approach can be useful if a long initiali-
zation sequence is required. The list of initialization words is
put into the buffer in the required order:
.VAR/DM/RAM/CIRC init_cmds[16]; {Codecinitsequence}
.VAR/DM/RAM stat_flag;
.INIT init_cmds:
b # 1 0 0 0 1 0 0 1 0 0 0 0 0 1 0 0 ,
b # 1 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 ,
b # 1 0 0 0 1 0 1 0 1 1 1 1 1 0 0 1 ,
b # 1 0 0 0 0 0 1 0 1 1 1 1 1 0 0 1 ,
b # 1 0 0 0 1 0 1 1 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 1 1 0 1 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 0 1 0 1 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 ,
b # 1 0 0 0 1 0 0 0 0 0 0 1 0 0 0 1 ,
b # 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 ;
In FSLB configurations, a single control word per codec per
sample period is sent to the AD73322L. In nonFSLB, it is
possible to initialize the device in a single sample period
provided the SCLK rate is programmed to a high rate. It is also
possible to use autobuffering, in which case an interrupt is
generated when the entire initialization sequence has been sent
to the AD73322L.
RUNNING THE AD73322L WITH ADCS OR DACS IN
POWER-DOWN
The programmability of the AD73322L allows the user flexi-
bility in choosing what sections of the AD73322L need to be
powered up. This allows better matching of the power con-
sumption and application requirements, because the AD73322L
offers two ADCs and two DACs in any combination. The
AD73322L always interfaces to the DSP in a standard way,
regardless of what ADC or DAC sections are enabled or
disabled. Therefore, the DSP expects to receive two ADC
samples per sample period and to transmit two DAC samples
per sample period. If a particular ADC is disabled (in power-
down) then its sample value is invalid. Likewise, a sample sent
to a DAC which is disabled has no effect.
The DSP program initializes pointers to the top of the buffer:
i3 = ^init_cmds;
13 = %init_cmds;
and puts the first entry in the DSP’s transmit buffer so that it is
available at the first SDOFS pulse:
ax0 = dm(i3,m1);
tx0 = ax0;
The DSP’s transmit interrupt is enabled:
imask = b#0001000000;
There are two distinct phases of operation of the AD73322L:
initialization of the device via each codec section’s control
registers, and operation of the converter sections of each codec.
The initialization phase involves programming the control
registers of the AD73322L to ensure the required operating
characteristics such as sampling rate, serial clock rate, and I/O
gain. There are several ways in which the DSP can be
programmed to initialize the AD73322L. These range from
hard-coding a sequence of DSP SPORT Tx register writes with
constants used for the initialization words, to putting the
initialization sequence in a circular data buffer and using an
autobuffered transmit sequence.
Rev. A | Page 38 of 48
AD73322L
At each occurrence of an SDOFS pulse, the DSP’s transmit
buffer contents are sent to the SDI pin of the AD73322L. This
also causes a subsequent DSP Tx interrupt which transfers the
initialization word, pointed to by the circular buffer pointer, to
the Tx buffer. The buffer pointer is updated to point to the next
unsent initialization word. When the circular buffer pointer
wraps around, which happens after the last word has been
accessed, it indicates that the initialization phase is complete.
This can be done manually in the DSP using a simple address
check, or autobuffered mode can be used to complete the
transfer automatically.
Because the AD73322L is effectively a cascade of two codec
units, it is important to observe the following restrictions in the
sequence of sending initialization words to the two codecs. It is
preferable to send pairs of control words for the corresponding
control registers in each codec, and it is essential to send the
control word for codec 2 before that for codec 1. Control
Registers A and B contain settings, such as sampling rate, serial
clock rate, etc., which critically require synchronous update in
both codecs.
Once the device has been initialized, Control Register A on
both codecs is written with a control word which changes the
operating mode from program mode to either data mode or
mixed control data mode. The device count field, which
defaults to 000b, must be programmed to 001b for a single
AD73322L device. In data mode or mixed mode, the main
function of the device is to return ADC samples from both
codecs and to accept DAC words for both codecs. During each
sample interval, two ADC samples are returned from the
device, while in the same interval two DAC update samples are
sent to the device. To reduce the number of interrupts and to
reduce complexity, autobuffering can be used to ensure that
only one interrupt is generated during each sampling interval.
txcdat: ar = dm(stat_flag);
ar = pass ar;
if eq rti;
ena sec_reg;
ax0 = dm (i3, m1);
tx0 = ax0;
ax0 = i3;
ay0 = ^init_cmds;
ar = ax0 - ay0;
if gt rti;
ax0 = 0x00;
dm (stat_flag) = ax0;
rti;
In the main body of the program the code loops, waiting for the
initialization sequence to be completed.
check_init:
ax0 = dm (stat_flag);
af = pass ax0;
if ne jump check_init;
Rev. A | Page 39 of 48
AD73322L
DAC TIMING CONTROL EXAMPLE
The AD73322L’s DAC is loaded from the DAC register contents
just before the ADC register contents are loaded to the serial
register (SDOFS going high). This default DAC load position
can be advanced in time to occur earlier with respect to the
SDOFS going high. Figure 50 shows an example of the ADC
unload and DAC load sequence. At time t1, the SDOFS is raised
to indicate that a new ADC word is ready. Following the
SDOFS pulse, 16 bits of ADC data are clocked out on SDO in
the subsequent 16 SCLK cycles, finishing at time t2 where the
DSP’s SPORT has received the 16-bit word. The DSP may
process this information and generate a DAC word to be sent to
the AD73322L. Time t3 marks the beginning of the sequence of
sending the DAC word to the AD73322L. This sequence ends at
time t4, where the DAC register is updated from the 16 bits in
the AD73322L’s serial register. However, the DAC is not
updated from the DAC register until time t5, which may not be
accept-able in certain applications. In order to reduce this delay
and load the DAC at time t6, the DAC advance register can be
programmed with a suitable setting corresponding to the
required time advance (refer to Table 15 for details of DAC
timing control settings).
SE
SCLK
SDOFS
SDO
ADC WORD
SDIFS
SDI
DAC WORD
DATA REGISTER
UPDATE
DAC LOAD
FROM DAC REGISTER
t1
t2
t3
t4 t6
t5
Figure 48. DAC Timing Control
Rev. A | Page 40 of 48
AD73322L
CONFIGURING AN AD73322L TO OPERATE IN DATA MODE
This section describes the typical sequence of control words
that are required to be sent to an AD73322L to set it up for data
mode operation.1 In this sequence, Registers B, C, and A are
programmed before the device enters data mode. This
description refers to the steps in Table 27.
Steps 7 to 9 are similar to Steps 1 to 3, but the user must
program Control Register A, with a device count field equal
to two channels in cascade, and set the PGM/DATA bit to
one to put the channel in data mode.
By Step 10, the programming phase completed, and actual
channel data read and write can begin. The words loaded in the
serial registers of the two channels at the ADC sampling event
contain valid ADC data, and the words written to the channels
from the DSP’s Tx register are interpreted as DAC words. The
DSP Tx register contains the DAC word for Channel 2.
At each sampling event, a pair of SDOFS pulses is observed,
which causes a pair of control (programming) words to be sent
to the device from the DSP. Each pair of control words should
program a single register in each Channel. The sequence to be
followed is Channel 2 followed by Channel 1.
Step 1 shows the first output sample event following a device
reset. The SDOFS signal is raised on both channels2 simulta-
neously, which prepares the DSP Rx register to accept the ADC
word from Channel 2, while SDOFS from Channel 1 becomes
an SDIFS to Channel 2. As the SDOFS of Channel 2 is coupled
to the DSP’s TFS and RFS, and to the SDIFS of Channel 1, this
event also forces a new control word to be output from the DSP
Tx register to Channel 1.3
In Step 11, the first DAC word has been transmitted into the
cascade, and the ADC word from Channel 2 has been read from
the cascade. The DSP Tx register contains the DAC word for
Channel 1. Because the words being sent to the cascade are
being interpreted as 16-bit DAC words, the addressing scheme
changes from one where the address was embedded in the
transmitted word, to one where the serial port counts the SDIFS
pulses. When the number of SDIFS pulses received equals the
value in the channel count field of Control Register A—the
length of the cascade—each channel updates its DAC register
with the present word in its serial register.
Step 2 shows the status of the channels following the transmis-
sion of the first control word. The DSP has received the output
word from Channel 2, while Channel 2 has received the output
word from Channel 1. Channel 1 has received the control word
destined for Channel 2. At this stage, the SDOFS of both
channels are again raised because Channel 2 has received
Channel 1’s output word, and as it is not a valid control word
addressed to Channel 2, it is passed on to the DSP. Likewise,
Channel 1 has received a control word destined for Channel 2—
address field is not zero—and it decrements the address field of
the control word and passes it on.
In Step 11 each channel has received only one SDIFS pulse;
Channel 2 received one SDIFS from the SDOFS of Channel 1
when it sent its ADC word, and Channel 1 received one SDIFS
pulse when it received the DAC word for Channel 2 from the
DSP’s Tx register. Therefore, each channel raises its SDOFS line
to pass on the current word in its serial register, and each
channel receives another SDIFS pulse.
Step 12 shows the completion of an ADC read and DAC write
cycle. Following Step 11, each channel has received two SDIFS
pulses that equal the setting of the channel count field in
Control Register A. The DAC register in each channel is up-
dated with the contents of the word that accompanied the
SDIFS pulse that satisfied the channel count requirement. The
internal frame sync counter is reset to zero and begins counting
for the next DAC update cycle.
Step 3 shows completion of the first series of control word
writes. The DSP has received both output words and each
channel has received a control word that addresses Control
Register B and sets the internal MCLK divider ratio to 1, SCLK
rate to DMCLK/2, and sampling rate to DMCLK/256. Both
channels are updated simultaneously because both receive the
addressed control word at the same time. This is an important
factor in cascaded operation as any latency between updating
the SCLK or DMCLK of channels can result in corrupted
operation. This does not happen in the case of an FSLB config-
uration, as shown here, but must be taken into account in a
nonFSLB configuration. Another observation of this sequence
is that the data-words are received and transmitted in reverse
order—that is, the ADC words are received by the DSP, Channel
2 first, then Channel 1 and, similarly, the transmit words from
the DSP are sent to Channel 2 first, then to Channel 1. This
ensures that all channels are updated at the same time.
Steps 10–12 are repeated on each sampling event.
1 Channel 1 and Channel 2 refer to the two AFE sections of the AD73322L.
2 The AD73322L is configured as two channels in cascade. The internal
cascade connections between Channels 1 and 2 are detailed in Figure 23.
The connections SDI/SDIFS are inputs to Channel 1, while SDO/SDOFS are
outputs from Channel 2.
3 This sequence assumes that the DSP SPORT’s Rx and Tx interrupts are
enabled. Ensure that there is no latency (separation) between control words
in a cascade configuration. This is especially the case when programming
Control Registers A and B as they must be updated synchronously in each
channel.
Steps 4 to 6 are similar to Steps 1–3, but the user must program
Control Register C to power up the analog sections of the
device (ADCs, DACs, and reference).
Rev. A | Page 41 of 48
AD73322L
Table 27. Data Mode Operation
Step
DSP Tx
AD73322L Channel 1
Data-word OUTPUT CH1 ->
0000000000000000
AD73322L Channel 2
Data-word OUTPUT CH2 ->
0000000000000000
DSPRx
1
Control Word CRB–CH2 ->
1000100100001011
Control Word CRB–CH1 ->
1000000100001011
2
3
Control Word CRB–CH2 ->
1000100100001011
Data-word OUTPUT CH1 ->
0000000000000000
Data-word OUTPUT CH2
0000000000000000
Control Word CRB–CH1
1000000100001011
Control Word CRB–CH2
1000000100001011
Data-word OUTPUT CH1
0000000000000000
At this time, Control Register B of both Channel 1 and Channel 2 are updated.
4
5
6
Control Word CRC–CH2 ->
1000101011111001
Data-word OUTPUT CH1 ->
0000000000000000
Data-word OUTPUT CH2 ->
0000000000000000
Control Word CRC–CH1 ->
1000001011111001
Control Word CRC–CH2 ->
1000101011111001
Data-word OUTPUT CH1 ->
0000000000000000
Data-word OUTPUT CH2
0000000000000000
Control Word CRC–CH1
1000001011111001
Control Word CRC–CH2
1000001011111001
Data-word OUTPUT CH1
0000000000000000
At this time, Control Register C of both Channel 1 and Channel 2 are updated.
7
8
9
Control Word CRA–CH2 ->
1000100000010001
Data-word OUTPUT CH1 ->
0000000000000000
Data-word OUTPUT CH2 ->
0000000000000000
Control Word CRA–CH1 ->
1000000000010001
Control Word CRA–CH2 ->
1000100000010001
Data-word OUTPUT CH1 ->
0000000000000000
Data-word OUTPUT CH2
0000000000000000
Control Word CRA–CH1
1000000000010001
Control Word CRA–CH2
1000000000010001
Data-word OUTPUT CH1
0000000000000000
At this time, Control Register A of both Channel 1 and Channel 2 are updated.
10
11
12
DAC WORD CH 2 ->
0111111111111111
DAC WORD CH 1 ->
1000000000000000
ADC Result CH1 ->
Unknown Data
ADC Result CH2 ->
Unknown Data
DAC Word CH 2 ->
0111111111111111
DAC Word CH 1
ADC Result CH1 ->
Unknown Data
ADC Result CH2
Unknown Data
ADC Result CH1
Unknown Data
DAC Word CH 2
0111111111111111
1000000000000000
At this time, the DAC of both Channel 1 and Channel 2 is updated and the ADC of both Channel 1 and Channel 2 has been read.
Rev. A | Page 42 of 48
AD73322L
CONFIGURING AN AD73322L TO OPERATE IN MIXED MODE
This section describes a typical sequence of control words that
would be sent to an AD73322L to configure it for operation in
mixed mode.1 It is not intended to be a definitive initialization
sequence, but shows users the typical input/output events that
occur in the programming and operation phases2. The text in
this section refers to the steps in Table 28.
subsequent interrupt service routine, the Tx register is loaded
with the control word for Channel 2. In Steps 9–10, Channels 1
and 2 are loaded with a control word setting for Control
Register B, which programs DMCLK = MCLK, the sampling
rate, to DMCLK/256, SCLK = DMCLK/2.
Steps 11 to 17 are similar to Steps 6 to 12 except that Control
Register C is programmed to power up all analog sections
(ADC, DAC, Reference = 1.2 V, REFOUT). In Steps 16–17,
DAC words are sent to the device—both DAC words are
necessary because each channel only updates its DAC when the
device has counted a number of SDIFS pulses, accompanied by
DAC words (in mixed mode, the MSB = 0), that are equal to the
device count field of Control Register A4. Because the channels
are in mixed mode, the serial port interrogates the MSB of the
16-bit word sent to determine whether it contains DAC data or
control information. DAC words should be sent in the sequence
Channel 2 followed by Channel 1.
Steps 1–5 detail the transfer of the control words to Control
Register A, which programs the device for mixed-mode
operation. Step 1 shows the first output sample event following
a device reset. The SDOFS signal is simultaneously raised on
both channels, which prepares the DSP Rx register to accept the
ADC word from Channel 2, while SDOFS from Channel 1
becomes an SDIFS to Channel 2. The cascade is configured as
nonFSLB, which means that the DSP has control over what is
transmitted to the cascade3 and, in this case, does not transmit
to the devices until both output words have been received from
the AD73322L.
Step 2 shows the status of the channels following receipt of the
Channel 2 output word. The DSP has received the ADC word
from Channel 2, while Channel 2 has received the output word
from Channel 1. At this stage, the SDOFS of Channel 2 is again
raised because Channel 2 has received Channel 1’s output word
and, as it is not addressed to Channel 2, passes it on to the DSP.
Steps 11 to17 show the control register update and DAC update
in a single sample period. Note that this combination is not
possible in the FSLB configuration3.
Steps 18 to 25 illustrate a control register readback cycle. In
Step 22, both channels have received a control word that
addresses Control Register C for readback (Bit 14 of the control
word = 1). When the channels receive the readback request, the
register contents are loaded to the serial registers, as shown in
Step 23. SDOFS is raised in both channels, which causes these
readback words to be shifted out toward the DSP. In Step 24,
the DSP has received the Channel 2 readback word, while
Channel 2 has received the Channel 1 readback word (note that
the address field in both words has been decremented to 111b).
In Step 25, the DSP has received the Channel 1 readback word
(its address field has been further decremented to 110b).
In Step 3, the DSP has received both ADC words. Typically, an
interrupt is generated following reception of the two output
words by the DSP (this involves programming the DSP to use
autobuffered transfers of two words). The transmit register of
the DSP is loaded with the control word destined for Channel 2.
This generates a transmit frame-sync (TFS) that is input to the
SDIFS input of the AD73322L to indicate the start of
transmission.
In Step 4, Channel 1 contains the control word destined for
Channel 2. The address field is decremented, SDOFS1 is raised
(internally) and the control word is passed on to Channel 2. The
Tx register of the DSP has now been updated with the control
word destined for Channel 1 (this can be done using auto-
buffering of transmit or by handling transmit interrupts
following each word sent).
Steps 26 to 30 detail an ADC and DAC update cycle using the
nonFSLB configuration. In this case, no control register update
is required.
1 Channel 1 and Channel 2 refer to the two AFE sections of the AD73322L.
2 This sequence assumes that the DSP SPORT’s Rx and Tx interrupts are
enabled. Ensure there is no latency (separation) between control words in a
cascade configuration. This is especially the case when programming
Control Registers A and B.
In Step 5, each channel has received a control word that
addresses Control Register A, sets the device count field equal
to two channels, and programs the channels into mixed mode
3 Mixed-mode operation with the FSLB configuration is more restricted in that
the number of words sent to the cascade equals the number of channels in
the cascade. This means that DAC updates may need to be substituted with
a register write or read. Using the FSLB configuration introduces a corruption
of the ADC samples in the sample period following a control register write.
This corruption is predictable and can be corrected in the DSP. The ADC
word is treated as a control word and the device address field is
decremented in each channel that it passes through before being returned
to the DSP.
(MM and
/DATA set to one).
PMG
Following Step 5, the device has been programmed into mixed
mode although none of the analog sections have been powered
up (controlled by Control Register C). Steps 6 to 10 detail
update of Control Register B in mixed mode. In Steps 6 to 8, the
ADC samples, which are invalid because the ADC section is not
yet powered up, are transferred to the DSP’s Rx section. In the
4 In mixed mode, DAC update is done using the same SDIFS counting scheme
as in normal data mode, with the exception that only DAC words (MSB set to
zero) are recognized as being able to increment the frame sync counters.
Rev. A | Page 43 of 48
AD73322L
Table 28. Mixed Mode Operation
Step
DSP Tx
AD73322L Channel 1
OUTPUT CH1 ->
AD73322L Channel 2
OUTPUT CH2 ->
DSP Rx
1
0000000000000000
0000000000000000
OUTPUT CH1 ->
2
3
4
5
OUTPUT CH2
0000000000000000
OUTPUT CH1
0000000000000000
CRA-CH2 ->
1000100000010011
CRA-CH1 ->
0000000000000000
CRA-CH2 ->
1000100000010011
CRA-CH1
1000000000010011
CRA-CH2
1000000000010011
1000000000010011
Control Register A of both channels has been programmed.
6
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2 ->
Unknown Data
7
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2
Unknown Data
ADC RESULT CH1
Unknown Data
8
CRB-CH2 ->
1000100100001011
CRB-CH1 ->
9
CRB-CH2 ->
1000000100001011
1000100100001011
CRB-CH1
10
CRB-CH2
1000000100001011
1000000100001011
The ADC data from both channels has been read and Control Register B of both channels has been programmed.
11
12
13
14
15
16
17
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2 ->
Unknown Data
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2
Unknown Data
ADC RESULT CH1
Unknown Data
CRC-CH2 ->
1000101011111001
CRC-CH1 ->
1000001011111001
DAC WORD CH 2 ->
0111111111111111
DAC WORD CH 1 ->
1000000000000000
CRC-CH2 ->
1000101011111001
CRC-CH1
1000001011111001
DAC WORD CH 2 ->
0111111111111111
DAC WORD CH 1
1000000000000000
CRC-CH2
1000001011111001
DAC WORD CH 2
0111111111111111
The ADC data from both channels has been read, Control Register C of both channels has been programmed, and DAC data for
both channels has been written.
18
19
20
21
22
23
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2 ->
Unknown Data
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2
Unknown Data
ADC RESULT CH1
Unknown Data
CRC-CH2 ->
11001010xxxxxxxx
CRC-CH1 ->
CRC-CH2 ->
11001010xxxxxxxx
CRC-CH1
10000010xxxxxxxx
READBACK CH 1 ->
1100001011111001
10000010xxxxxxxx
CRC-CH2
10000010xxxxxxxx
READBACK CH 2 ->
1100001011111001
Rev. A | Page 44 of 48
AD73322L
Step
DSP Tx
AD73322L Channel 1
AD73322L Channel 2
READBACK CH 1 ->
1111101011111001
DSP Rx
24
READBACK CH 2
1111101011111001
READBACK CH 1
25
1111001011111001
The ADC data of both channels has been read, and a readback of Control Register C has been performed.
26
27
28
29
30
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2 ->
Unknown Data
ADC RESULT CH1 ->
Unknown Data
ADC RESULT CH2
Unknown data
ADC RESULT CH1
Unknown Data
DAC WORD CH 2 ->
0111111111111111
DAC WORD CH 1 ->
1000000000000000
DAC WORD CH 2 ->
0111111111111111
DAC WORD CH 1
DAC WORD CH 2
1000000000000000
0111111111111111
The ADC data from both channels has been read, and the DAC data for both channels has been written.
Rev. A | Page 45 of 48
AD73322L
OUTLINE DIMENSIONS
18.10 (0.7126)
17.70 (0.6969)
28
1
15
14
7.60 (0.2992)
7.40 (0.2913)
10.65 (0.4193)
10.00 (0.3937)
2.65 (0.1043)
2.35 (0.0925)
0.75 (0.0295)
0.25 (0.0098)
× 45°
0.30 (0.0118)
0.10 (0.0039)
8°
0°
1.27 (0.0500)
SEATING
PLANE
0.51 (0.0201)
0.31 (0.0122)
1.27 (0.0500)
0.40 (0.0157)
0.33 (0.0130)
0.20 (0.0079)
COPLANARITY
0.10
BSC
COMPLIANT TO JEDEC STANDARDS MS-013AE
CONTROLLING DIMENSIONS ARE IN MILLIMETERS; INCH DIMENSIONS
(IN PARENTHESES) ARE ROUNDED-OFF MILLIMETER EQUIVALENTS FOR
REFERENCE ONLY AND ARE NOT APPROPRIATE FOR USE IN DESIGN
Figure 49. 28-Lead Standard Small Outline Package [SOIC]
Wide Body
(RW-28)
Dimensions shown in millimeters and (inches )
9.80
9.70
9.60
28
15
4.50
4.40
4.30
6.40 BSC
1
14
PIN 1
0.65
BSC
1.20 MAX
0.15
0.05
8°
0°
0.75
0.60
0.45
0.30
0.19
0.20
0.09
SEATING
PLANE
COPLANARITY
0.10
COMPLIANT TO JEDEC STANDARDS MO-153AE
Figure 50. 28-Lead Thin Shrink Small Outline Package [TSSOP]
(RU-28)
Dimensions shown in millimeters
0.75
0.60
0.45
1.60 MAX
16.00 BSC SQ
44
34
1
33
PIN 1
TOP VIEW
(PINS DOWN)
14.00
BSC SQ
10°
6°
2°
1.45
1.40
1.35
0.20
0.09
7°
VIEW A
11
23
3.5°
0°
12
22
0.15
0.05
SEATING
PLANE
0.50
0.42
0.35
1.00 BSC
LEAD PITCH
0.10 MAX
COPLANARITY
VIEW A
ROTATED 90° CCW
LEAD WIDTH
COMPLIANT TO JEDEC STANDARDS MS-026-BEA
Figure 51. 44-Lead Low Profile Quad Flat Package [LQFP]
(ST-44-2)
Dimensions shown in millimeters
Rev. A | Page 46 of 48
AD73322L
ORDERING GUIDE
Model
AD73322LAR
Temperature Range
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +105°C
−40°C to +105°C
−40°C to +105°C
−40°C to +105°C
−40°C to +105°C
−40°C to +105°C
−40°C to +105°C
Package Description
Wide Body SOIC
Package Option
RW-28
RW-28
RW-28
RU-28
RU-28
RU-28
RU-28
ST-44A
ST-44A
RW-28
RW-28
RW-28
RU-28
RU-28
ST-44A
ST-44A
AD73322LAR-REEL
AD73322LAR-REEL7
AD73322LARU
Wide Body SOIC
Wide Body SOIC
Thin Shrink TSSOP
AD73322LARU-REEL
AD73322LARUZ1
Thin Shrink TSSOP
Thin Shrink TSSOP
AD73322LARUZ-REEL
1
Thin Shrink TSSOP
AD73322LAST
Plastic Thin Quad Flatpack (LQFP)
Plastic Thin Quad Flatpack (LQFP)
Wide Body SOIC
AD73322LAST-REEL
AD73322LYR
AD73322LYR-REEL
AD73322LYR-REEL7
AD73322LYRU
Wide Body SOIC
Wide Body SOIC
Thin Shrink TSSOP
AD73322LYRU-REEL
AD73322LYST
Thin Shrink TSSOP
Plastic Thin Quad Flatpack (LQFP)
Plastic Thin Quad Flatpack (LQFP)
AD73322LYST-REEL
1 Z = Pb-free part.
Rev. A | Page 47 of 48
AD73322L
NOTES
©
2004 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
C00691–0–12/04(A)
Rev. A | Page 48 of 48
相关型号：
AD73322LYRZ
Dual-Channel, 3 V Front-End Processor for General Purpose Applications Including Speech and Telephony
ADI
AD73322LYSTZ-RL
IC SPECIALTY TELECOM CIRCUIT, PQFP44, PLASTIC, MS-026-BEA, LQFP-44, Telecom IC:Other
ADI
©2020 ICPDF网 联系我们和版权申明