AD AD73311LAR

a
Low Cost, Low Power CMOS
General Purpose Analog Front End
AD73311L
FEATURES
16-Bit A/D Converter
16-Bit D/A Converter
Programmable Input/Output Sample Rates
76 dB ADC SNR
77 dB DAC SNR
Programmable Sampling Rate
64 kS/s Maximum Sample Rate
–90 dB Crosstalk
Low Group Delay (25 ms Typ per ADC Channel,
50 ms Typ per DAC Channel)
Programmable Input/Output Gain
Flexible Serial Port Which Allows Up to Eight Devices
to Be Connected in Cascade
Single (+3 V) Supply Operation
33 mW Max Power Consumption at 2.7 V
On-Chip Reference
20-Lead SOIC/SSOP/TSSOP Packages
APPLICATIONS
General Purpose Analog I/O
Speech Processing
Cordless and Personal Communications
Telephony
Active Control of Sound and Vibration
Data Communications
GENERAL DESCRIPTION
The AD73311L is a complete front-end processor for general
purpose applications including speech and telephony. It features
a 16-bit A/D conversion channel and a 16-bit D/A conversion
channel. Each channel provides 70 dB signal-to-noise ratio over
a voiceband signal bandwidth. The final channel bandwidth can
be reduced, and signal-to-noise ratio improved, by external
digital filtering in a DSP engine.
The AD73311L is suitable for a variety of applications 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 gains of the A/D and D/A conversion channels are programmable over 38 dB and 21 dB ranges respectively. An on-chip
reference voltage is included to allow single supply operation.
A serial port (SPORT) allows easy interfacing of single or cascaded devices to industry standard DSP engines.
The AD73311L is available in 20-lead SOIC, SSOP and
TSSOP packages.
FUNCTIONAL BLOCK DIAGRAM
AVDD1
VINP
VINN
ANALOG
LOOPBACK/
SINGLE-ENDED
ENABLE
AVDD2
DVDD
SDI
ANALOG
SIGMA-DELTA
MODULATOR
0/38dB
PGA
DECIMATOR
SDIFS
SCLK
SERIAL
I/O
PORT
VOUTP
+6/–15dB
PGA
VOUTN
REFCAP
CONTINUOUS
TIME
LOW-PASS FILTER
SWITCHEDCAPACITOR
LOW-PASS FILTER
1-BIT
DAC
DIGITAL
SIGMA-DELTA
MODULATOR
INTERPOLATOR
SDO
SDOFS
SE
MCLK
RESET
REFERENCE
AD73311L
REFOUT
AGND1
AGND2
DGND
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
which may result from its use. No license is granted by implication or
otherwise under any patent or patent rights of Analog Devices.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781/329-4700
World Wide Web Site: http://www.analog.com
Fax: 781/326-8703
© Analog Devices, Inc., 2000
AD73311L–SPECIFICATIONS1 F = 8 kHz; T = T
(AVDD = DVDD = 2.7 V to 3.3 V; DGND = AGND = 0 V, fDMCLK = 16.384 MHz,
S
A
MIN to TMAX, unless otherwise noted.)
Parameter
REFERENCE
REFCAP
Absolute Voltage, VREFCAP
REFCAP TC
REFOUT
Typical Output Impedance
Absolute Voltage, VREFOUT
Minimum Load Resistance
Maximum Load Capacitance
Min
1.08
1.08
1
–0.6
–1.0
± 0.1
71
70
76
74
72
56
60
59
–20
Group Delay4, 5
Input Resistance at VIN 2, 4
DAC SPECIFICATIONS
Maximum Voltage Output Swing 2
Single-Ended
Differential
Nominal Voltage Output Swing (0 dBm0)
Single-Ended
Differential
PGA = 6 dB
Total Harmonic Distortion
PGA = 0 dB
PGA = 6 dB
Intermodulation Distortion
Idle Channel Noise
Crosstalk
145
1.2
–2.2
Total Harmonic Distortion
PGA = 0 dB
PGA = 38 dB
Intermodulation Distortion
Idle Channel Noise
Crosstalk
Output Bias Voltage4
Absolute Gain
Gain Tracking Error
Signal to (Noise + Distortion)
PGA = 0 dB
1.32
1.32
1.578
–2.85
1.0954
–6.02
PGA = 38 dB
DC Offset
Power Supply Rejection
1.2
50
100
ADC SPECIFICATIONS
Maximum Input Range at VIN 2, 3
Nominal Reference Level at VIN
(0 dBm0)
Absolute Gain
PGA = 0 dB
PGA = 38 dB
Gain Tracking Error
Signal to (Noise + Distortion)
PGA = 0 dB
AD73311LA
Typ
Max
1.08
–1.8
70
+1.0
Unit
V
ppm/°C
Ω
V
kΩ
pF
Test Conditions/Comments
0.1 µF Capacitor Required from
REFCAP to AGND2
Unloaded
V p-p
dBm
V p-p
dBm
Measured Differentially
Max Input = (1.578/1.2) × VREFCAP
Measured Differentially
dB
dB
dB
1.0 kHz, 0 dBm0
1.0 kHz, 0 dBm0
1.0 kHz, +3 dBm0 to –50 dBm0
Refer to Figure 5a
300 Hz to 3400 Hz
0 Hz to fSAMP/2
300 Hz to 3400 Hz; fSAMP = 64 kHz
0 Hz to fSAMP/2; fSAMP = 64 kHz
300 Hz to 3.4 kHz
0 Hz to fSAMP/2
dB
dB
dB
dB
dB
dB
–85
–85
–82
–76
–100
–75
dB
dB
dB
dBm0
dB
25
45
µs
kΩ6
300 Hz to 3.4 kHz
300 Hz to 3.4 kHz
PGA = 0 dB
PGA = 0 dB
ADC Input Signal Level: 1.0 kHz, 0 dBm0
DAC Input at Idle
PGA = 0 dB
Input Signal Level at AVDD and DVDD
Pins 1.0 kHz, 100 mV p-p Sine Wave
64 kHz Output Sample Rate
DMCLK = 16.384 MHz
+2
–84
+25
mV
dB
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
Max Output = 2 × ((1.578/1.2) × VREFCAP
1.0954
–6.02
2.1909
0
1.2
–0.7
± 0.1
V p-p
dBm
V p-p
dBm
V
dB
dB
PGA = 6 dB
1.32
+0.4
77
76
77
77
–80
–80
–76
–82
–100
dB
dB
dB
dB
–70
–2–
dB
dB
dB
dBm0
dB
PGA = 6 dB
REFOUT Unloaded
1.0 kHz, 0 dBm0
1.0 kHz, +3 dBm0 to –50 dBm0
Refer to Figure 5b
300 Hz to 3.4 kHz Frequency Range
300 Hz to 3400 Hz; fSAMP = 64 kHz
300 Hz to 3.4 kHz Frequency Range
300 Hz to 3400 Hz; fSAMP = 64 kHz
PGA = 0 dB
PGA = 0 dB
ADC Input Signal Level: AGND; DAC
Output Signal Level: 1.0 kHz, 0 dBm0
REV. A
AD73311L
Parameter
DAC SPECIFICATIONS (Continued)
Power Supply Rejection
Min
AD73311LA
Typ
Group Delay4, 5
Output DC Offset 2, 7
Minimum Load Resistance, R L2, 8
Single-Ended
Differential
Maximum Load Capacitance, CL2, 8
Single-Ended
Differential
FREQUENCY RESPONSE
(ADC AND DAC)9 Typical Output
0
0.03125
0.0625
0.125
0.1875
0.25
0.3125
0.375
0.4375
> 0.5
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
IDD10
–30
Max
Unit
Test Conditions/Comments
–81
dB
25
µs
Input Signal Level at AVDD and DVDD
Pins: 1.0 kHz, 100 mV p-p Sine Wave
64 kHz Input Sample Rate, Interpolator
Bypassed (CRE:5 = 1)
PGA = 6 dB
+5
+50
mV
Ω
Ω
150
150
500
100
pF
pF
Normalized to fSAMP
0
–0.1
–0.25
–0.6
–1.4
–2.8
–4.5
–7.0
–9.5
< –12.5
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
VDD – 0.8
0
VDD
0.8
10
10
V
V
µA
pF
VDD – 0.4
0
–10
VDD
0.4
+10
V
V
µA
2.7
2.7
3.3
3.3
V
V
Channel Frequency Response Is
Programmable by Means of External
Digital Filtering
|IOUT| ≤ 100 µA
|IOUT| ≤ 100 µA
See Table I
NOTES
1
Operating temperature range is as follows: –40°C to +105°C. Therefore, TMIN = –40°C and TMAX = +105°C.
2
Test conditions: Input PGA set for 0 dB gain, Output PGA set for 6 dB gain, no load on analog outputs (unless otherwise noted).
3
At input to sigma-delta modulator of ADC.
4
Guaranteed by design.
5
Overall group delay will be affected by the sample rate and the external digital filtering.
6
The ADC’s input impedance is inversely proportional to DMCLK and is approximated by: (4 × 1011)/DMCLK.
7
Between VOUTP and VOUTN.
8
At VOUT output.
9
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.
10
Test Conditions: no load on digital inputs, analog inputs ac coupled to ground, no load on analog outputs.
Specifications subject to change without notice.
Table I. Current Summary (AVDD = DVDD = 3.3 V)
Conditions
Analog Internal Digital External Interface
Current Current
Current
Total Current
(Max)
SE
MCLK
ON
Comments
ADC Only On
ADC and DAC On
REFCAP Only On
REFCAP and
REFOUT Only On
All Sections Off
2
5.6
0.65
4.5
4.8
0
0.5
0.5
0
8.0
12.5
1.0
1
1
0
YES
YES
NO
2.7
0
0
0.6
0
0
3.8
0.75
0
0
NO
YES
All Sections Off
1 µA
0.5 µA
0
20 µA
0
NO
The above values are in mA and are typical values unless otherwise noted.
REV. A
–3–
REFOUT Disabled
REFOUT Disabled
REFOUT Disabled
MCLK Active Levels Equal to
0 V and DVDD
Digital Inputs Static and Equal
to 0 V or DVDD
AD73311L
Table II. Signal Ranges
Parameter
Condition
VREFCAP
VREFOUT
ADC
Signal 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
Differential
Nominal Voltage
Output Swing
Single-Ended
Differential
Output Bias Voltage
DAC
TIMING CHARACTERISTICS
1.578 V p-p
3.156 V p-p
1.0954 V p-p
2.1909 V p-p
VREFOUT
(AVDD = DVDD = 2.7 V to 3.6 V; AGND = DGND = 0 V; TA = TMlN to TMAX, unless otherwise noted)
Parameter
Limit at
TA = –40ⴗC to +105ⴗC
Unit
Description
Clock Signals
t1
t2
t3
61
24.4
24.4
ns min
ns min
ns min
See Figure 1
MCLK Period
MCLK Width High
MCLK Width Low
Serial Port
t4
t5
t6
t7
t8
t9
t10
t11
t12
t13
t1
0.4 × t1
0.4 × t1
20
0
10
10
10
10
30
ns min
ns min
ns min
ns min
ns min
ns max
ns min
ns min
ns max
ns max
See Figures 3 and 4
SCLK Period
SCLK Width High
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
t1
100␮A
IOL
t2
TO OUTPUT
PIN
2.1V
CL
15pF
100␮A
t3
Figure 1. MCLK Timing
IOH
Figure 2. Load Circuit for Timing Specifications
t1
t2
t3
MCLK
t13
SCLK*
t5
t6
t4
* SCLK IS INDIVIDUALLY PROGRAMMABLE
IN FREQUENCY (MCLK/4 SHOWN HERE).
Figure 3. SCLK Timing
–4–
REV. A
AD73311L
SE (I)
SCLK (O)
THREESTATE
t7
SDIFS (I)
t8
t8
t7
SDI (I)
SDOFS (O)
SDO (O)
D15
t9
THREESTATE
D14
D15
D0
D1
t10
t11
t12
THREESTATE
D15
D2
D1
D15
D0
D14
80
80
70
70
60
60
50
50
S/(N+D) – dB
S/(N+D) – dB
Figure 4. Serial Port (SPORT)
40
30
30
20
20
10
10
0
0
–10
–85
–75
–65
–55
–45
–35
VIN – dBm0
–25
–15
–10
–85
–5 0
3.17
Figure 5a. S/(N+D) vs. VIN (ADC @ 3 V) over Voiceband
Bandwidth (300 Hz – 3.4 kHz)
REV. A
40
–75
–65
–55
–45 –35
VIN – dBm0
–25
–15
–5 0
3.17
Figure 5b. S/(N+D) vs. VIN (DAC @ 3 V) over Voiceband
Bandwidth (300 Hz – 3.4 kHz)
–5–
AD73311L
ABSOLUTE MAXIMUM RATINGS*
PIN CONFIGURATION
(TA = 25°C unless otherwise noted)
AVDD, DVDD to GND . . . . . . . . . . . . . . . –0.3 V to +4.6 V
AGND to DGND . . . . . . . . . . . . . . . . . . . . –0.3 V to +0.3 V
Digital I/O Voltage to DGND . . . –0.3 V to (DVDD + 0.3 V)
Analog I/O Voltage to AGND . . . –0.3 V to (AVDD + 0.3 V)
Operating Temperature Range
Industrial (A Version) . . . . . . . . . . . . . . . –40°C to +105°C
Storage Temperature Range . . . . . . . . . . . . –65°C to +150°C
Maximum Junction Temperature . . . . . . . . . . . . . . . . 150°C
SOIC, θJA Thermal Impedance . . . . . . . . . . . . . . . . . 75°C/W
Lead Temperature, Soldering
Vapor Phase (60 sec) . . . . . . . . . . . . . . . . . . . . . . . . 215°C
Infrared (15 sec) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220°C
SSOP, θJA Thermal Impedance . . . . . . . . . . . . . . . . 126°C/W
Lead Temperature, Soldering
Vapor Phase (60 sec) . . . . . . . . . . . . . . . . . . . . . . . . 215°C
Infrared (15 sec) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220°C
TSSOP, θJA Thermal Impedance . . . . . . . . . . . . . . . 143°C/W
Lead Temperature, Soldering
Vapor Phase (60 sec) . . . . . . . . . . . . . . . . . . . . . . . . 215°C
Infrared (15 sec) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220°C
VOUTP 1
20 SE
VOUTN 2
19 SDI
AVDD1 3
18 SDIFS
AGND1 4
17 SDOFS
VINP 5
AD73311L 16 SDO
VINN 6
TOP VIEW 15 MCLK
(Not to Scale)
14 SCLK
REFOUT 7
REFCAP 8
13 RESET
AVDD2 9
12 DVDD
AGND2 10
11 DGND
ORDERING GUIDE
*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.
Model
Temperature
Range
Package
Option1
AD73311LAR
AD73311LARS
AD73311LARU
EVAL-AD73311LEB
–40°C to +105°C
–40°C to +105°C
–40°C to +105°C
Evaluation Board2
R-20
RS-20
RU-20
NOTES
1
R = 0.3' Small Outline IC (SOIC), RS = Shrink Small Outline Package (SSOP),
RU = Thin Small Shrink Outline Package (TSSOP).
2
The AD73311L evaluation board features a cascade of two codecs interfaced to
an ADSP-2185L DSP. The board features a DSP software monitor which
allows interface to a PC serial port.
CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection.
Although the AD73311L 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.
–6–
WARNING!
ESD SENSITIVE DEVICE
REV. A
AD73311L
PIN FUNCTION DESCRIPTIONS
Pin
Number
Mnemonic
Function
1
2
3
4
5
6
7
8
VOUTP
VOUTN
AVDD1
AGND1
VINP
VINN
REFOUT
REFCAP
9
10
11
12
13
AVDD2
AGND2
DGND
DVDD
RESET
14
SCLK
15
16
MCLK
SDO
17
SDOFS
18
SDIFS
19
SDI
20
SE
Analog Output from the Positive Terminal of the Output Channel.
Analog Output from the Negative Terminal of the Output Channel.
Analog Power Supply Connection for the Output Driver.
Analog Ground Connection for the Output Driver.
Analog Input to the Positive Terminal of the Input Channel.
Analog Input to the Negative Terminal of the Input Channel.
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 Connection.
Digital Ground/Substrate Connection.
Digital Power Supply Connection.
Active Low Reset Signal. This input resets the entire chip, resetting the control registers and clearing
the digital circuitry.
Output Serial Clock whose 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.
Master Clock Input. MCLK is driven from an external clock signal.
Serial Data Output of the Codec. 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.
Framing Signal Output for SDO Serial Transfers. The frame sync is on 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 ignored when SE is low.
Serial Data Input of the Codec. 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.
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.
REV. A
–7–
AD73311L
TERMINOLOGY
Absolute Gain
ABBREVIATIONS
ADC
Analog-to-Digital Converter.
ALB
Analog Loop-Back.
BW
Bandwidth.
CRx
A Control Register where x is a placeholder for an
alphabetic character (A–E). There are five read/
write control registers on the AD73311L—designated CRA through CRE.
CRx:n
A bit position, where n is a placeholder for a
numeric character (0–7), within a control register;
where x is a placeholder for an alphabetic character (A–E). Position 7 represents the MSB and
Position 0 represents the LSB.
DAC
Digital-to-Analog Converter.
DLB
Digital Loop-Back.
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.
FSLB
Idle channel noise is defined as the total signal energy measured
at the output of the device when the input is grounded (measured
in the frequency range 300 Hz–3400 Hz).
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.
PGA
Programmable Gain Amplifier.
Intermodulation Distortion
SC
Switched Capacitor.
With inputs consisting of sine waves at two frequencies, fa and
fb, any active device with nonlinearities will create 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 are 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).
SNR
Signal-to-Noise Ratio.
SPORT
Serial Port.
THD
Total Harmonic Distortion.
VBW
Voice Bandwidth.
Absolute gain is 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.
Crosstalk
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.
Gain Tracking Error
Gain tracking error 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.
Group Delay
Group delay is defined as the derivative of radian phase with
respect to radian frequency, dø(f)/df. Group delay is a measure
of 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.
Idle Channel Noise
Power Supply Rejection
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).
Sample Rate
The sample rate is the rate at which the ADC updates its output register and the DAC updates its output from its input
register. It is fixed relative to the DMCLK (= DMCLK/256)
and therefore may only be changed by changing the DMCLK.
SNR+THD
Signal-to-noise ratio plus harmonic distortion is defined to be
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–3400 Hz, including harmonics but excluding dc.
–8–
REV. A
AD73311L
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 6a). 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 6b). The combination of these techniques, followed by the application of a
digital filter, reduces the noise in band sufficiently to ensure
good dynamic performance from the part (Figure 6c).
FUNCTIONAL DESCRIPTION
Encoder Channel
The encoder channel consists of an input configuration block, 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.
Input Configuration Block
The input configuration block consists of a multiplexing arrangement that allows selection of various input configurations. This
includes ADC input selection from either the VINP, VINN pins
or from the DAC output via the Analog Loop-Back (ALB)
arrangement. Differential inputs can be inverted and it is also
possible to use the device in single-ended mode, which allows
the option of using the VINP, VINN pins as two separate
single-ended inputs, either of which can be selected under
software control.
BAND
OF
INTEREST
FS/2
DMCLK/16
a.
Programmable Gain Amplifier
The encoder section’s analog front end comprises a switched
capacitor PGA that 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 III, 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.
NOISE-SHAPING
BAND
OF
INTEREST
FS/2
DMCLK/16
b.
The PGA gain is set by bits IGS0, IGS1 and IGS2 (CRD:0–2)
in Control Register D.
DIGITAL FILTER
Table III. PGA Settings for the Encoder Channel
IGS2
IGS1
IGS0
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
0
6
12
18
20
26
32
38
BAND
OF
INTEREST
c.
Figure 6. Sigma-Delta Noise Reduction
Figure 7 shows the various stages of filtering that are employed
in a typical AD73311L application. In Figure 7a we see 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 bandwidth of interest. In Figure
7b, the signal and noise-shaping responses of the sigma-delta
modulator are shown. The signal response provides further
rejection of any high frequency signals while the noise-shaping
will push the inherent quantization noise to an out-of-band
position. The detail of Figure 7c shows the response of the
digital decimation filter (Sinc-cubed response) with nulls every
multiple of DMCLK/256, which is the decimation filter update
rate. The final detail in Figure 7d 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 Figures 7a through 7c is implemented in the AD73311L.
ADC
The ADC consists 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 bitstream, representing the analog input
signal, is input to the antialiasing decimation filter. The decimation filter reduces the sample rate and increases the resolution.
Analog Sigma-Delta Modulator
The AD73311L input channel employs 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 highest
frequency of interest. In the case of the AD73311L, the initial
REV. A
FS/2
DMCLK/16
–9–
AD73311L
ADC Coding
FB = 4kHz
The ADC coding scheme is in twos complement format (see
Figure 8). 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 15-bit word, which is the 16-bit
transfer being used as a flag bit to indicate either control or data
in the frame.
FSINIT = DMCLK/8
a. Analog Antialias Filter Transfer Function
VINN
VREF + (VREF ⴛ 0.32875)
SIGNAL TRANSFER FUNCTION
ANALOG
INPUT
VREF
VREF – (VREF ⴛ 0.32875)
NOISE TRANSFER FUNCTION
VINP
10...00
FB = 4kHz
FSINIT = DMCLK/8
00...00
01...11
ADC CODE DIFFERENTIAL
b. Analog Sigma-Delta Modulator Transfer Function
VREF + (VREF ⴛ 0.6575)
VINN
ANALOG
INPUT
VREF
VREF – (VREF ⴛ 0.6575)
FB = 4kHz FSINTER = DMCLK/256
VINP
10...00
c. Digital Decimator Transfer Function
00...00
01...11
ADC CODE SINGLE-ENDED
Figure 8. ADC Transfer Function
Decoder Channel
The decoder channel consists of a digital interpolator, digital
sigma-delta modulator, a single bit digital-to-analog converter
(DAC), an analog smoothing filter and a programmable gain
amplifier with differential output.
FB = 4kHz FSFINAL = 8kHz
DAC Coding
FSINTER = DMCLK/256
The DAC coding scheme is in twos complement format with
0x7FFF being full-scale positive and 0x8000 being full-scale
negative.
d. Final Filter LPF (HPF) Transfer Function
Figure 7. AD73311L ADC Frequency Responses
Decimation Filter
The digital filter used in the AD73311L carries out two important functions. Firstly, it removes the out-of-band quantization
noise, which is shaped by the analog modulator and secondly, it
decimates the high frequency bitstream to a lower rate 15-bit word.
The antialiasing decimation filter is a sinc-cubed digital filter
that reduces the sampling rate from DMCLK/8 at the modulator to an output rate at the SPORT of DMCLK/M (where M
depends on the sample rate setting—M = 256 @ 64 kHz; M =
512 @ 32 kHz, M = 1024 @ 16 kHz, M = 2048 @ 8 kHz), and
increases the resolution from a single bit to 15 bits. Its Z transform is given as: [(1–Z–N)/(1–Z–1)]3 where N is determined by
the sampling rate (N = 32 @ 64 kHz, N = 64 @ 32 kHz, N =
128 @ 16 kHz, N = 256 @ 8 kHz). This ensures a minimal
group delay of 25 µs at the 64 kHz sampling rate.
Interpolation Filter
The anti-imaging interpolation filter is a sinc-cubed digital filter
which up-samples the 16-bit input words from the SPORT
input rate of DMCLK/M (where M depends on the sample rate
setting—M = 256 @ 64 kHz; M = 512 @ 32 kHz, M = 1024 @
16 kHz, M = 2048 @ 8 kHz), to a rate of DMCLK/8 while
filtering to attenuate images produced by the interpolation process. Its Z transform is given as: [(1–Z–N)/(1–Z–1)]3 where N is
determined by the sampling rate (N = 32 @ 64 kHz, N = 64 @
32 kHz, N = 128 @ 16 kHz, N = 256 @ 8 kHz). The DAC
receives 16-bit samples from the host DSP processor at a rate of
DMCLK/M. 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.
–10–
REV. A
AD73311L
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 passband of the converter. The bitstream output of the
sigma-delta modulator is fed to the single bit DAC where it is
converted to an analog voltage.
Analog Smoothing Filter and 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). 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 IV. The PGA gain is
set by bits OGS0, OGS1 and OGS2 (CRD:4-6) in Control
Register D.
Table IV. 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
–3
–6
–9
–12
–15
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.
Voltage Reference
SPORT Overview
The AD73311L SPORT is a flexible, full-duplex, synchronous
serial port whose protocol has been designed to allow up to eight
AD73311L devices to be connected, in cascade, to a single DSP
via a six-wire interface. It has a very flexible architecture that can
be configured by programming two of the internal control registers. The AD73311L SPORT has three distinct modes of operation: Control Mode, Data Mode and Mixed Control/Data Mode.
In Control Mode (CRA:0 = 0), the device’s internal configuration can be programmed by writing to the five internal control
registers. In this mode, control information can be written to or
read from the codec. In Data Mode (CRA:0 = 1), information
that is 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 is written to or
read from the device. Mixed mode (CRA:0 = 1 and CRA:1 = 1)
allows the user to choose whether the information being sent to
the device contains either 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.
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 there are some precautions
that must be observed. The primary precaution is that no information must be written to the SPORT without reference to an
output sample event, which is when the serial register will be
overwritten with the latest ADC sample word. Once the SPORT
starts to output the latest ADC word then it is safe for the DSP
to write new control or data words to the codec. In certain configurations, data can be written to the device to coincide with
the output sample being shifted out of the serial register—see
section on interfacing devices. 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 will happen.
The codec communicates 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 SPORT block diagram, shown in Figure 9, details the six
control registers (A–F), external MCLK to internal DMCLK
divider and serial clock divider. The divider rates are controlled
by the setting of Control Register B. The AD73311L 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. 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.
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 uses a common serial register for
serial input and output, communications between an AD73311L
codec and a host processor (DSP engine) must always be initiated by the codec itself. This ensures that there is no danger of
the information being sent to the codec being corrupted by
ADC samples being output by the codec.
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. When working at the lower SCLK rate of
DMCLK/8, which is intended for interfacing with slower DSPs,
the SPORT will support a maximum of two devices in cascade
with the sample rate of DMCLK/256.
The AD73311L reference, REFCAP, is a bandgap 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.
The reference output (REFOUT) can be enabled for biasing
external circuitry by setting the RU bit (CRC:6) of CRC.
Serial Port (SPORT)
REV. A
–11–
AD73311L
MCLK
(EXTERNAL)
MCLK
DIVIDER
DMCLK
(INTERNAL)
3
SCLK
SCLK
DIVIDER
SERIAL PORT
(SPORT)
SE
RESET
SDOFS
SDIFS
SDI
SDO
SERIAL REGISTER
REGISTER
SERIAL
2
8
CONTROL
REGISTER A
8
8
8
CONTROL
REGISTER B
8
8
CONTROL
REGISTER D
CONTROL
REGISTER C
CONTROL
REGISTER E
CONTROL
REGISTER F
Figure 9. SPORT Block Diagram
SPORT Register Maps
There are two register banks for the AD73311L: the control
register bank and the data register bank. The control register
bank consists of six read/write registers, each eight bits wide.
Table IX shows the control register map for the AD73311L.
The first two control registers, CRA and CRB, are reserved for
controlling the SPORT. They hold settings for parameters such
as bit rate, internal master clock rate and device count (used
when more than one AD73311L is connected in cascade from
a single SPORT). The other three registers; CRC, CRD and
CRE are used to hold control settings for the ADC, DAC,
Reference and Power Control sections of the device. 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.
Master Clock Divider
The AD73311L features a programmable master clock divider
that allows the user to reduce an externally available master
clock, at pin MCLK, by one of the ratios 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 V shows
the division ratio corresponding to the various bit settings. The
default divider ratio is divide-by-one.
is programmable by setting bits CRB:2–3. Table VI shows the
serial clock rate corresponding to the various bit settings.
Table VI. SCLK Rate Divider Settings
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 AD73311L 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
SCD0
SCLK Rate
0
0
1
1
0
1
0
1
DMCLK/8
DMCLK/4
DMCLK/2
DMCLK
Sample Rate Divider
The AD73311L features a programmable sample rate divider
that allows users flexibility in matching the codec’s ADC and
DAC sample 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 VII shows the sample rate corresponding to the various
bit settings.
Table VII. Sample Rate Divider Settings
Table V. DMCLK (Internal) Rate Divider Settings
MCD2
SCD1
DIR1
DIR0
SCLK Rate
0
0
1
1
0
1
0
1
DMCLK/2048
DMCLK/1024
DMCLK/512
DMCLK/256
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/(DMCLK/8); see Table VIII. In certain
circumstances this can reduce the group delay when the ADC
and DAC are used to process data in series. Appendix E details
how the DAC advance feature can be used.
NOTE: The DAC advance register should be changed before
the DAC section is powered up.
–12–
REV. A
AD73311L
Table VIII. DAC Timing Control
DA4
DA3
DA2
DA1
DA0
Time Advance*
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
0 ns
488.2 ns
976.5 ns
—
14.64 µs
15.13 µs
that the device must be programmed to the correct settings after
power-up or reset. Following a reset, the SDOFS will be asserted
2048 DMCLK cycles after RESET going high. 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.
Power Management
*DMCLK = 16.384 MHz.
OPERATION
Resetting the AD73311L
The pin RESET resets all the control registers. All registers are
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 RESET pin, the
device can be reset using the RESET bit (CRA:7) in Control
Register A. Both hardware and software resets require 4 DMCLK
cycles. On reset, DATA/PGM (CRA:0) is set to 0 (default condition) thus enabling Program Mode. The reset conditions ensure
The individual functional blocks of the AD73311L 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 their design. The power
control register provides individual control settings for the major
functional blocks 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
will still be enabled, in this case, because its individual bit is set.
Refer to Table XIII for details of the settings of CRC.
Table IX. Control Register Map
Address (Binary)
Name
Description
Type
Width
Reset Setting (Hex)
000
001
010
011
100
101
110 to 111
CRA
CRB
CRC
CRD
CRE
CRF
Control Register A
Control Register B
Control Register C
Control Register D
Control Register E
Control Register F
Reserved
R/W
R/W
R/W
R/W
R/W
R/W
8
8
8
8
8
8
0x00
0x00
0x00
0x00
0x00
0x00
Table X. Control Word Description
15
14
C/D
R/W
13
12
11
Device Address
10
9
8
7
Register Address
6
5
4
3
2
1
0
Register Data
Control
Frame
Description
Bit 15
Control/Data
When set high, it 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, it 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 13–11
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 10–8
Register Address
This 3-bit field is used to select one of the five control registers on the AD73311L.
Bits 7–0
Register Data
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
–13–
AD73311L
Table XI. Control Register A Description
CONTROL REGISTER A
7
RESET
6
DC2
5
DC1
4
DC0
3
SLB
2
DLB
1
MM
0
DATA/PGM
Bit
Name
Description
0
1
2
3
4
5
6
7
DATA/PGM
MM
DLB
SLB
DC0
DC1
DC2
RESET
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)
Device Count (Bit 1)
Device Count (Bit 2)
Software Reset (0 = Off; 1 = Initiates Reset)
Table XII. Control Register B Description
CONTROL REGISTER B
7
6
5
4
3
2
1
0
CEE
MCD2
MCD1
MCD0
SCD1
SCD0
DIR1
DIR0
Bit
Name
Description
0
1
2
3
4
5
6
7
DIR0
DIR1
SCD0
SCD1
MCD0
MCD1
MCD2
CEE
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)
Table XIII. Control Register C Description
CONTROL REGISTER C
7
6
5
4
3
2
1
0
–
RU
PUREF
PUDAC
PUADC
–
–
PU
Bit
Name
Description
0
1
2
3
4
5
6
PU
Reserved
Reserved
PUADC
PUDAC
PUREF
RU
7
Reserved
Power-Up Device (0 = Power Down; 1 = Power On)
Must Be Programmed to Zero (0)
Must Be Programmed to Zero (0)
ADC Power (0 = Power Down; 1 = Power On)
DAC Power (0 = Power Down; 1 = Power On)
REF Power (0 = Power Down; 1 = Power On)
REFOUT Use (0 = Disable REFOUT; 1 = Enable
REFOUT)
Must Be Programmed to Zero (0)
–14–
REV. A
AD73311L
Table XIV. Control Register D Description
CONTROL REGISTER D
7
6
5
MUTE
OGS2
OGS1
4
3
OGS0 RMOD
2
1
0
IGS2
IGS1
IGS0
Bit
Name
Description
0
1
2
3
4
5
6
7
IGS0
IGS1
IGS2
RMOD
OGS0
OGS1
OGS2
MUTE
Input Gain Select (Bit 0)
Input Gain Select (Bit 1)
Input Gain Select (Bit 2)
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)
Table XV. Control Register E Description
CONTROL REGISTER E
7
6
5
4
3
2
1
0
–
–
IBYP
DA4
DA3
DA2
DA1
DA0
Bit
Name
Description
0
1
2
3
4
5
DA0
DA1
DA2
DA3
DA4
IBYP
6
7
Reserved
Reserved
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)
Must Be Programmed to Zero (0)
Must Be Programmed to Zero (0)
Table XVI. Control Register F Description
CONTROL REGISTER F
REV. A
7
6
5
4
3
2
1
0
ALB
INV
SEEN
–
–
–
–
–
Bit
Name
Description
0
1
2
3
4
5
6
7
Reserved
Reserved
Reserved
Reserved
Reserved
SEEN
INV
ALB
Must Be Programmed to Zero (0)
Must Be Programmed to Zero (0)
Must Be Programmed to Zero (0)
Must Be Programmed to Zero (0)
Must Be Programmed to Zero (0)
Single-Ended Enable (0 = Disabled; 1 = Enabled)
Input Invert (0 = Disabled; 1 = Enabled)
Analog Loopback of Output to Input (0 = Disabled; 1 = Enabled)
–15–
AD73311L
There are five operating modes available on the AD73311L.
Two of these—Digital Loop-Back and Sport Loop-Back—are
provided as diagnostic modes with the other three, Program,
Data and Mixed Program/Data, being available for general
purpose use. 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 will
be programmed for 16-bit transfers.
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 (i.e., mixed
mode disabled), it must receive a hardware reset to reprogram
any of the control register settings. In a single codec configuration, each 16-bit data frame sent from the DSP to the device is
interpreted as DAC data. The default device count is 1, therefore
each input frame sync will cause the 16-bit data frame to be
loaded to the DAC register.
Program (Control) Mode
Mixed Program/Data Mode
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 X. 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)
then 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. 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). In a single codec
configuration, all control word addresses must be zero, otherwise they will not be recognized; in a multi-codec configuration
all addresses from zero to N-1 (where N = number of devices in
cascade) are valid.
This mode allows the user to send control words to the device
along with the DAC data. This permits adaptive control of the
device whereby control of the input/output gains can be effected
by interleaving control words along with the normal flow of
DAC data. The standard data frame remains 16 bits, but now
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 DATA/PGM
bit (CRA:0) to 1. In the case where control setting changes will
be 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.
Operating Modes
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 10, or they can lag the output words by a time interval
that should not exceed the sample interval. After reset, output
frame sync pulses will occur at a slower default sample rate, which
is DMCLK/2048, until Control Register B is programmed after
which the SDOFS pulses will occur at a rate set by the DIR0-1 bits
of CRB. This is to allow slow controller devices to establish
communication with the AD73311L. During Program Mode,
the data output by the device is random and should not be interpreted as ADC data.
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 by programming the DATA/PGM (CRA:0) bit to a 1 and MM (CRA:1) to
0. Once the device is in Data Mode, the 16-bit input data frame
is now interpreted as DAC data rather than a control frame. This
data is therefore loaded directly to the DAC register. In Data
Mode, as the entire input data frame contains DAC data, the
device relies on counting the number of input frame syncs
Digital Loop-Back
This mode can be used for diagnostic purposes and allows 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 will continue to work,
which allows the user to control gain settings, etc. Only when
DLB is enabled with Mixed Mode operation can the user disable
the DLB, otherwise the device must be reset.
Sport Loop-Back
This mode allows the user to verify the DSP interfacing and
connection by writing words to the SPORT of the device and
have them returned back unchanged at the next sample interval.
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.
Analog Loop-Back
In Analog Loop-Back mode, the differential DAC output is
connected, via a loop-back switch, to the ADC input (see Figure
12). 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).
–16–
REV. A
AD73311L
SE
SCLK
SDOFS
SDO
SAMPLE WORD (DEVICE 1)
SAMPLE WORD (DEVICE 1)
DATA (CONTROL) WORD (DEVICE 1)
DATA (CONTROL) WORD (DEVICE 1)
SDIFS
SDI
Figure 10. Interface Signal Timing for Single Device Operation
SE
SCLK
SDOFS(2)
SDO(2)
SAMPLE WORD (DEVICE 2)
SAMPLE WORD (DEVICE 1)
SDOFS(1)
SDIFS(2)
SDO(1)
SDI(2)
SAMPLE WORD (DEVICE 1)
DATA (CONTROL) WORD (DEVICE 2)
SDIFS(1)
SDI(1)
DATA (CONTROL) WORD (DEVICE 2)
DATA (CONTROL) WORD (DEVICE 1)
Figure 11. Interface Signal Timing for Cascade of Two Devices
REV. A
–17–
AD73311L
ANALOG
LOOP-BACK
SELECT
inputs as the codec SDOFS will be input to both. This configuration guarantees that input and output events occur simultaneously
and is the simplest configuration for operation in normal Data
Mode. Note that 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 will be transmitted to coincide with the
first output word from the device(s).
SINGLEENDED
ENABLE
INVERT
VINP
0/38dB
PGA
VINN
VREF
SDIFS
TFS
SDI
DT
ADSP-218x
DSP
VOUTP
+6/–15dB
PGA
VOUTN
SCLK
CONTINUOUS
TIME
LOW-PASS
FILTER
AD73311L
CODEC
SCLK
DR
SDO
RFS
SDOFS
AD73311L
REFOUT
REFERENCE
Figure 13. Indirectly Coupled or Nonframe Sync LoopBack Configuration
REFCAP
Cascade Operation
Figure 12. Analog Loop-Back Connectivity
INTERFACING
The AD73311L 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 accompanying frame synchronization signal which 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 13, 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. 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 codec(s). This means that full control can be implemented over the device configuration as well as updating the
DAC in a given sample interval. The second configuration
(shown in Figure 14) 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 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
The AD73311L has been designed to support up to eight codecs
in a cascade connected to a single serial port (see Figure 37).
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.
There may be some restrictions in cascade operation due to the
number of devices configured in the cascade and the serial clock
rate chosen. Table XVII details the requirements for SCLK rate
for cascade lengths from 1 to 8 devices. This assumes a directly
coupled frame sync arrangement as shown in Figure 13.
Table XVII. Cascade Options
SCLK
1
DMCLK
DMCLK/2
DMCLK/4
DMCLK/8
✓
✓
✓
✓
–18–
Number of Devices in Cascade
2
3
4
5
6
7
✓
✓
✓
✓
TFS
DT
ADSP-218x
DSP
SCLK
DR
RFS
✓
✓
✓
X
✓
✓
✓
X
✓
✓
X
X
✓
✓
X
X
✓
✓
X
X
8
✓
✓
X
X
SDIFS
SDI
SCLK
AD73311L
CODEC
SDO
SDOFS
Figure 14. Directly Coupled or Frame Sync LoopBack Configuration
REV. A
AD73311L
When using the indirectly coupled frame sync configuration in
cascaded operation it is necessary to be aware of the restrictions
in sending data to all devices in the cascade. Effectively the time
allowed is given by the sampling interval (256/DMCLK) which
is 15.625 µs for a sample rate of 64 kHz. In this interval, the
DSP must transfer N × 16 bits of information where N is the
number of devices in the cascade. 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:
256/DMCLK > ((N × 16/SCLK) + TINTERRUPT LATENCY)
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.
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 cascade (see Table XVIII). However, for cascade
operation this field must contain a binary value that is one less
than the number of devices in the cascade.
The range of sampling rates is aimed to offer the user a degree
of flexibility in deciding how their 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 15 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, it is possible to eliminate
the noise floor 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 16
shows the resulting spectrum following the filtering and decimation of the spectrum of Figure 15 from 64 kHz to an 8 kHz rate.
0
–20
–40
dB
–60
Table XVIII. Device Count Settings
–80
DC2
DC1
DC0
Cascade Length
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
1
2
3
4
5
6
7
8
–100
–120
–140
0
0.5
1.0
1.5
2.0
FREQUENCY – Hz
2.5
3.0
3.5
ⴛ 104
Figure 15. FFT (ADC 64 kHz Sampling)
0
PERFORMANCE
–20
As the AD73311L is designed to provide high performance, low
cost conversion, it is important to understand the means by
which this high performance can be achieved in a typical application. This section will, by means of spectral graphs, outline the
typical performance of the device and highlight 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.
dB
–40
–60
–80
–100
Encoder Section
The AD73311L 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 below will show the
performance of a 1 kHz sine wave when converted at the various
sample rates.
REV. A
–120
0
500
1000
1500 2000 2500
FREQUENCY – Hz
3000
3500
4000
Figure 16. FFT (ADC 8 kHz Filtered and Decimated from
64 kHz)
–19–
AD73311L
The AD73311L 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 17 details the spectrum of a 1 kHz test tone converted
at an 8 kHz rate.
Figure 19 details the spectrum of the final 8 kHz sampled
filtered tone.
0
–20
–40
0
dB
–60
–80
50
dB
–100
–120
100
150
–140
500
1000
1500
2000
2500
FREQUENCY Hz
3000
3500
4000
Figure 19. FFT (ADC 8 kHz Filtered and Decimated from
16 kHz)
0
500
1000
1500
2000
2500
FREQUENCY – Hz
3000
3500
4000
Encoder Group Delay
Figure 17. FFT (ADC 8 kHz Direct Sampling)
The device features an on-chip master clock divider circuit that
allows the sample rate to be reduced as the sampling rate of the
sigma-delta converter is proportional to the output of the MCLK
Divider (whose default state is divide by 1).
When programmed for high sampling rates, the AD73311L
offers a very low level of group delay, which is given by the
following relationship:
Group Delay (Decimator) = Order × ((M – 1)/2) × TDEC
where:
Order is the order of the decimator (= 3),
The decimator’s frequency response (Sinc3) gives some passband 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.
In the case of voiceband processing where 4 kHz represents the
Nyquist frequency, if the signal to be measured were externally
bandlimited, an 8 kHz sampling rate would suffice. However, if
it is required to limit the bandwidth using a digital filter, 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 bandlimited signal at an 8 kHz
rate. Figure 18 details the initial 16 kHz sampled tone.
0
–20
–40
dB
–60
–80
–100
–120
–140
0
0
1000
2000
3000
4000
5000
FREQUENCY – Hz
6000
7000
M is the decimation factor (= 32 @ 64 kHz, = 64 @ 32 kHz,
= 128 @ 16 kHz , = 256 @ 8 kHz) and
TDEC is the decimation sample interval (= 1/2.048e6) (based
on DMCLK = 16.384 MHz) => Group Delay (Decimator @
64 kHz) = 3 × (32 – 1)/2 × (1/2.048e6) = 22.7 µs
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 20 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 voiceband SNR, the user can implement an initial
anti-imaging filter, preceded by 8 kHz to 64 kHz interpolation,
in the DSP.
8000
Figure 18. FFT (ADC 16 kHz Direct Sampling)
–20–
REV. A
0
0
–10
–10
–20
–20
–30
–30
–40
–40
dB
dB
AD73311L
–50
–60
–60
–70
–70
–80
–80
–90
–90
–100
0
–100
0
0.5
1.0
1.5
2.0
FREQUENCY – Hz
2.5
3.0
3.5
0.5
ⴛ 104
As the AD73311L can be operated at 8 kHz (see Figure 21) or
16 kHz sampling rates, which make it particularly suited for
voiceband processing, it is important to 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 will be
generated at integer intervals of the sampling frequency. In this
case these images must be removed by external analog filtering.
1.0
1.5
2.0
FREQUENCY – Hz
2.5
3.0
3.5
ⴛ 104
Figure 22. FFT (DAC 8 kHz Sampling—Interpolator
Bypassed)
Figure 20. FFT (DAC 64 kHz Sampling)
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) and
TINT is the interpolation sample interval (= 1/2.048e6)
=> Group Delay (Interpolator @ 64 kHz)
0
dB
–50
–10
= 3 × (32 – 1)/2 × (1/2.048e6)
–20
= 22.7 µs
–30
The analog section has a group delay of approximately 25 µs.
–40
On-Chip Filtering
–50
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 DMCLK
of 16.384 MHz) and the simple, external RC antialias filter is
sufficient to provide the required stopband 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 passband without the application of some further antialias filtering.
In the AD73311L, 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.
–60
–70
–80
–90
–100
0
500
1000
1500
2000
2500
FREQUENCY – Hz
3000
3500
4000
Figure 21. FFT (DAC 8 kHz Sampling)
Figure 22 shows the output spectrum of a 1 kHz tone being
generated at an 8 kHz sampling rate with the interpolator
bypassed.
REV. A
–21–
AD73311L
DESIGN CONSIDERATIONS
20
The AD73311L features both differential inputs and outputs on
each channel to provide optimal performance and avoid commonmode 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.
I/O CHANNEL GAIN
0
–20
–40
ANTIALIAS
FILTER
–60
100⍀
–80
VINP
0.047␮F
0/38dB
PGA
0.047␮F
–100
0
1
2
3
4
5
6
VINN
100⍀
7
VREF
ⴛ 104
FREQUENCY – Hz
Figure 23. Codec Uncompensated Input-to-Output
Frequency Response (fSAMP = 64 kHz)
VOUTP
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.
VOUTN
REFOUT
I/O CHANNEL GAIN
0
–20
–40
–60
–80
–100
0
1
2
3
4
FREQUENCY – Hz
5
6
7
ⴛ 104
Figure 24. Codec Compensated Input-to-Output
Frequency Response (fSAMP = 64 kHz)
CONTINUOUS
TIME
LOW-PASS
FILTER
REFERENCE
AD73311L
REFCAP
0.1␮F
Figure 25. Analog Input (DC-Coupled)
The spectral response of a sinc-cubed filter shows the characteristic nulls at integer intervals of the sampling frequency. Its
passband 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 will
not give sufficient attenuation of frequencies outside the band of
interest, therefore, it may be necessary to implement a final filter
in the DSP which will equalize the passband rolloff and provide
a sharper transition band and greater stopband attenuation.
20
+6/–15dB
PGA
Analog Inputs
The analog input (encoder) section of the AD73311L can be
interfaced to external circuitry in either ac-coupled or dc-coupled
modes.
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 singleended inputs connected to the positive and negative input pins.
The primary concerns in interfacing to the ADC are firstly to
provide adequate antialias filtering and to ensure that the signal
source will drive the switched-capacitor input of the ADC correctly. The sigma-delta design of the ADC and its oversampling
characteristics simplify the antialias requirements but it must be
remembered that 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. It is recommended that for optimum performance
the capacitors used for the antialiasing filter be of high quality
dielectric (NPO). The second issue mentioned above 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, it is important to understand that the slew
rate characteristic is an important consideration when choosing
external buffers for use with the AD73311L. The internal inverting
op amps on the AD73311L are specifically designed to interface
to the ADC’s SC input stage.
The AD73311L’s on-chip 38 dB preamplifier can be enabled
when there is not enough gain in the input circuit; the preamplifier 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.
–22–
REV. A
AD73311L
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), 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 in Figures 27 through 29.
If the ADC is being connected in single-ended mode, the
AD73311L should be programmed for single-ended mode using
the SEEN and INV bits of CRF, and the inputs connected as
shown in Figure 28. When operated in single-ended input mode,
the AD73311L can multiplex one of the two inputs to the ADC
input, as shown in Figures 28 and 29.
0.1␮F 100⍀
VINP
0.047
␮F
VINN
VINP
10k⍀
0/38dB
PGA
VINN
0/38dB
PGA
VREF
VREF
VOUTP
VOUTP
VOUTN
OPTIONAL
BUFFER
+6/15dB
PGA
CONTINUOUS
TIME
LOW-PASS
FILTER
VOUTN
+6/–15dB
PGA
CONTINUOUS
TIME
LOW-PASS
FILTER
REFOUT
REFOUT
REFERENCE
REFERENCE
AD73311L
AD73311L
REFCAP
REFCAP
0.1␮F
0.1␮F
Figure 28. Analog Input (AC-Coupled) Single-Ended
Figure 26. Analog Input (DC-Coupled) Using External
Amplifiers
The AD73311L’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 dcor 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 25, 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 26.
VINP
0.1␮F
10k⍀
100⍀
0/38dB
PGA
VINN
0.047
␮F
VREF
VOUTP
VOUTN
+6/15dB
PGA
CONTINUOUS
TIME
LOW-PASS
FILTER
REFOUT
REFERENCE
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 27.
AD73311L
REFCAP
0.1␮F
Figure 29. Analog Input (AC-Coupled) Single-Ended
(Alternate Input)
Interfacing to an Electret Microphone
0.1␮F
0.1␮F
100⍀
10k⍀
0.047␮F
0/38dB
PGA
VINN
100⍀
10k⍀
VINP
0.047␮F
VREF
VOUTP
VOUTN
+6/–15dB
PGA
CONTINUOUS
TIME
LOW-PASS
FILTER
REFOUT
REFERENCE
REFCAP
AD73311L
0.1␮F
Figure 30 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 that supplies power to the microphone; therefore,
this output signal must be capacitively coupled to remove the
power supply (dc) component. In this circuit the AD73311L
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, the microphone can be powered from the
system power supply.
Figure 27. Analog Input (AC-Coupled) Differential
REV. A
–23–
AD73311L
Figure 32 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.
5V
RA
C1
10␮F
R2
RB
R1
VINP
C2
0/38dB
PGA
VINN
ELECTRET
MICROPHONE
VINP
VREF
VOUTP
+6/–15dB
PGA
VOUTN
VINN
CONTINUOUS
TIME
LOW-PASS
FILTER
REFOUT
REFERENCE
COUT
VOUTP
+6/–15dB
PGA
RLOAD
AD73311L
VOUTN
CONTINUOUS
TIME
LOW-PASS
FILTER
REFCAP
REFOUT
CREFCAP
AD73311L
REFERENCE
REFCAP
Figure 30. Electret Microphone Interface Circuit
CREFCAP
Analog Output
The AD73311L’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 that
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 31 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:
COUT =
Figure 32. Example Circuit for Single-Ended Output
Differential-to-Single-Ended Output
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 33 shows a scheme for doing this.
VINP
VINN
1
2 π fC RLOAD
RF
VOUTP
RI
RLOAD
where fC = desired cutoff frequency.
VOUTN
RF
RI
+6/–15dB
PGA
CONTINUOUS
TIME
LOW-PASS
FILTER
REFOUT
REFERENCE
AD73311L
REFCAP
CREFCAP
VINP
Figure 33. Example Circuit for Differential-to-SingleEnded Output Conversion
VINN
Digital Interfacing
COUT
VOUTP
RLOAD
VOUTN
+6/–15dB
PGA
CONTINUOUS
TIME
LOW-PASS
FILTER
COUT
REFOUT
REFERENCE
AD73311L
REFCAP
CREFCAP
Figure 31. Example Circuit for Differential Output
The AD73311L is designed to easily interface to most common
DSPs. The SCLK, SDO, SDOFS, SDI and SDIFS must be
connected to the SCLK, DR, RFS, DT and TFS pins of the
DSP respectively. The SE pin may be controlled from a parallel
output pin or flag pin such as FL0–2 on the ADSP-218x (or XF
on the TMS320C5x) or, where SPORT power-down is not
required, it can be permanently strapped high using a suitable
pull-up resistor. The RESET pin may be connected to the 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 necessary to operate the device in mixed mode,
which allows a software reset, otherwise there is no convenient
way of resetting the device. Figures 34 and 35 show typical
connections to an ADSP-218x and TMS320C5x respectively.
–24–
REV. A
AD73311L
SDIFS
TFS
SDI
DT
SCLK
SCLK
ADSP-218x
DSP
complete the cascade. SE and RESET on all devices are fed
from the signals that were synchronized with the MCLK using
the circuit as described above. The SCLK from only one device
need be connected to the DSP’s SCLK input(s) as all devices
will be running at the same SCLK frequency and phase.
DR
SDO
RFS
AD73311L
CODEC
TFS
DT
SDOFS
FL0
RESET
FL1
SE
SDIFS
ADSP-218x
DSP
SDI
SCLK
SCLK
DR
FL0
SE
AD73311L
CODEC
SDOFS
DEVICE 1
FL1
SDIFS
FSX
SDIFS
SCLK
SCLK
CLKX
TMS320C5x
DSP
DR
SE
AD73311L
CODEC
RESET
SDO
AD73311L
CODEC
CLKR
MCLK
SDI
SDI
DT
RESET
SDO
RFS
Figure 34. AD73311L Connected to ADSP-218x
MCLK
SDOFS
DEVICE 2
SDO
FSR
SDOFS
XF
RESET
D1
D2
Q1
74HC74
Q2
SE
Figure 37. Connection of Two AD73311Ls Cascaded to
ADSP-218x
Figure 35. AD73311L Connected to TMS320C5x
Cascade Operation
Grounding and Layout
Where it is required to configure a cascade of up to eight
devices, it is necessary to ensure that the timing of the SE and
RESET signals are synchronized at each device in the cascade.
A simple D-type flip-flop is sufficient to sync each signal to the
master clock MCLK, as in Figure 36.
DSP CONTROL
TO SE
D
Q
Since the analog inputs to the AD73311L are differential, most
of the voltages in the analog modulator are common-mode
voltages. The excellent common-mode rejection of the part will
remove common-mode noise on these inputs. The analog and
digital supplies of the AD73311L 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 AD73311’s ADC is
high, and the noise levels from the AD73311L are so low, care
must be taken with regard to grounding and layout.
SE SIGNAL SYNCHRONIZED
TO MCLK
1/2
74HC74
MCLK
CLK
DSP CONTROL
TO RESET
D
Q
RESET SIGNAL SYNCHRONIZED
TO MCLK
1/2
74HC74
MCLK
CLK
Figure 36. SE and RESET Sync Circuit for Cascaded
Operation
Connection of a cascade of devices to a DSP, as shown in Figure 37, 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
REV. A
The printed circuit board that houses the AD73311L should be
designed so the analog and digital sections are separated and
confined to certain sections of the board. The AD73311L 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 38. A minimum etch technique
is generally best for ground planes as 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
to use a ferrite bead inductor as shown in Figure 38.
–25–
AD73311L
On ADSP-218x processors, it is necessary to enable SPORT
interrupts and use Interrupt Service Routines (ISRs) to handle
Tx/Rx activity, while on the TMS320C5x processors it is possible to poll the status of the Rx and Tx registers, which means
that Rx/Tx activity can be monitored using a single ISR that
would ideally be the Tx ISR as the Tx interrupt will typically
occur before the Rx ISR.
DIGITAL GROUND
ANALOG GROUND
DSP SOFTWARE CONSIDERATIONS WHEN
INTERFACING TO THE AD73311L
Figure 38. Ground Plane Layout
Avoid running digital lines under the device for they will couple
noise onto the die. The analog ground plane should be allowed
to run under the AD73311L to avoid noise coupling. The power
supply lines to the AD73311L 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, and 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 will
reduce the effects of feedthrough through the board. A microstrip
technique is by far the best 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.
Good decoupling is important when using high speed devices.
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. To achieve the best
from these decoupling capacitors, they should be placed as close
as possible to the device, ideally right up against it. In systems
where a common supply voltage is used to drive both the AVDD
and DVDD of the AD73311, 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 AD73311L and AGND and the recommended digital supply decoupling capacitors between the DVDD pin and DGND.
DSP Programming Considerations
This section discusses some aspects of 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 AD73311L:
•
•
•
•
•
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.
• Frame Syncs occur one SCLK cycle before the MSB of the
• serial word.
• Frame Syncs are active high.
DSP SPORT Interrupts
It is important when choosing the operating mode and hardware
configuration of the AD73311L to 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. As most modern DSP engines
support these modes, this appendix will attempt to discuss these
topics in a generic DSP sense.
Operating Mode
The AD73311L supports two basic operating modes: Frame Sync
Loop Back (FSLB) and nonFSLB (see 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. 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 AD73311L outputs two
sample words per sample period, it is possible using autobuffering
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 AD73311L) while also
having some status flags to indicate where each new sample
comes from (or is destined for).
Mixed-Mode Operation
To take full advantage of Mixed-Mode operation, it is necessary
to configure the DSP/Codec interface in nonFSLB and to disable
autobuffering. This allows a variable numbers of words to be sent
to the AD73311L in each sample period—the extra words being
control words which 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.
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 which may cause some discontinuity in the output signal due to a sample point being missed
and the previous sample being repeated. This however may be
acceptable in some cases as the effect may be masked by gain
changes, etc.
If SPORT interrupts are enabled, it is important to note that the
active signals on the frame sync pins do not necessarily correspond with the positions in time of where SPORT interrupts are
generated.
–26–
REV. A
AD73311L
Interrupts
The AD73311L 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, it is important to note that 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.
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 below.
ax0
= b#1000000100000100;
tx0
= ax0;
idle; {wait for tx register to send current word}
The circular buffer approach can be useful if a long initialization
sequence is required. The list of initialization words is put into
the buffer in the required order.
.VAR/DM/RAM/CIRC init_cmds[8]; {Codec init sequence}
.VAR/DM/RAM stat_flag;
.INIT init_cmds:
b#1000000100000100,
b#1000001011111001,
b#1000001100000000,
b#1000010000000000,
b#1000010100000000,
b#1000011000000000,
b#1000011100000000,
b#1000000000010001,
Initialization
Following reset, the AD73311L 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. As
communications between AD73311L and the DSP are interrupt
driven, it is usually not practical to embed the initialization 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.
In FSLB configurations, a single control word per codec per
sample period is sent to the AD73311L whereas 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 AD73311L.
Running the AD73311L with ADCs or DACs in Power-Down
The programmability of the AD73311L allows the user flexibility
in choosing which sections of the AD73311L need be powered
up. This allows better matching of the power consumption to the
application requirements as the AD73311L offers an ADC and
a DAC in any combination. The AD73311L always interfaces to
the DSP in a standard way regardless of whether the ADC or
DAC sections are enabled or disabled. Therefore, the DSP will
expect to receive an ADC samples per sample period and to
transmit two DAC samples per sample period. If the ADC is
disabled (in power-down), its sample value will be invalid. Likewise, a sample sent to a DAC that is disabled will have no effect.
There are two distinct phases of operation of the AD73311L:
initialization of the device via the control registers, and operation
of the converter sections of each codec. The initialization phase
involves programming the control registers of the AD73311L to
ensure the required operating characteristics such as sampling
rate, serial clock rate, I/O gain, etc. There are several ways in
which the DSP can be programmed to initialize the AD73311L.
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
and the DSP program initializes pointers to the top of the buffer
i3 = ^init_cmds;
l3 = %init_cmds;
and puts the first entry in the DSP’s transmit buffer so it is
available at the first SDOFS pulse.
ax0 = dm(i3,m1);
tx0 = ax0;
The DSP’s transmit interrupt is enabled.
imask = b#0001000000;
At each occurrence of an SDOFS pulse, the DSP’s transmit
buffer contents are sent to the SDI pin of the AD73311L. 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 the complete transfer automatically.
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;
–27–
AD73311L
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;
If the AD73311L is used in a cascade of two or more codec
units, it is important to observe some restrictions in the sequence
of sending initialization words to the two codecs. It is preferable
to send groups of control words for the corresponding control
registers in each codec and it is essential to send the control
words in descending order—last device first . . . first device last.
Control Registers A and B contain settings, such as sampling
rate, serial clock rate etc., which critically require synchronous
update in each codec.
Once the device has been initialized, Control Register A on each
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, will
have to be programmed to the required setting—depending on the
number of devices in cascade. In Data Mode or Mixed Mode,
the main function of the device is to return ADC samples from
each codec and to accept DAC words for each codecs. During
each sample interval, ADC samples will be returned from the
device equal to the number of devices in cascade, while in the
same interval DAC update samples will be sent to the device
again the number of DAC words being equal to the number of
devices in the cascade.
In order 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.
–28–
REV. A
AD73311L
APPENDIX A
Configuring an AD73311L to Operate in Data Mode
This section describes the typical sequence of control words that
are required to be sent to an AD73311L to set it up for data mode
operation. In this sequence, Registers B, C and A are programmed
before the device enters data mode. This description panel refers
to Table XIX.
At each sampling event, an SDOFS pulse will be observed that
will cause a control (programming) word to be sent to the device
from the DSP.
Step 1: The first output sample event following device reset.
The SDOFS signal is raised, which prepares the DSP Rx
Register to accept the ADC word from the AD73311L. As the
AD73311L’s SDOFS is coupled to the DSP’s TFS and RFS, and
to the SDIFS of the AD73311L, this event also forces a new control word to be output from the DSP Tx Register to the AD73311L.
Step 2: We observe the status of the channel following the
transmission of the control word. The DSP has received the
ADC word (invalid because the ADC is not yet powered up)
from the AD73311L and the AD73311L has received the control
word destined for Control Register B. At this stage, the eight LSBs
of the control word are loaded to Control Register B, which sets
the internal MCLK divider ratio to 1, SCLK rate to DMCLK/8.
Step 3: The next ADC sample event that occurs raises the
SDOFS line of the AD73311L. The DSP Tx Register contains
the control word to be written to the AD73311L.
Step 4: Following transmission of the control word, the DSP
Rx Register contains the ADC word that during Program Mode
is a copy of the control word written at the previous sampling
interval where the device address field (Bits 13–11) have been
decremented from 000b to 111b. The AD73311L has received a
control word that is addressed to Control Register C, which turns
on power to the ADC, DAC, REFCAP and buffered REFOUT.
Steps 5 and 6: The programming phase is completed by sending a control word addressed to Control Register A, which sets
the device in Data Mode.
Step 7: The AD73311L provides its first valid ADC sample as
the ADC has been powered up and data mode is enabled. In
data mode all words sent to the device are interpreted as DAC
words. Likewise, all words received from the device are interpreted as ADC words.
Step 8: The first DAC word has been transmitted to the device
and is loaded to the internal DAC register.
Steps 9 and 10: Another ADC read and DAC write cycle.
Table XIX.
Step
DSP Tx
AD73311L
DSP Rx
1
2
1000000100001011
1000001011111001
0000000000000000
1000000100001011
xxxxxxxxxxxxxxxx
0000000000000000
3
4
1000001011111001
1000000000000001
1000000100001011
1000001001111001
xxxxxxxxxxxxxxxx
1011100100001011
5
6
1000000000000001
DAC WORD N
1000001011111001
1000000000000001
xxxxxxxxxxxxxxxx
1011101011111001
7
8
DAC WORD N
DAC WORD N+1
ADC RESULT N
DAC WORD N
xxxxxxxxxxxxxxxx
ADC RESULT N
9
10
DAC WORD N+1
DAC WORD N+2
ADC RESULT N+1
DAC WORD N+1
xxxxxxxxxxxxxxxx
ADC RESULT N+1
REV. A
–29–
AD73311L
APPENDIX B
Configuring an AD73311L to Operate in Mixed Mode
the Tx register is loaded with the control word setting for Control Register B which programs DMCLK = MCLK, the sampling rate to DMCLK/256, SCLK = DMCLK/2.
This section describes a typical sequence of control words that
would be sent to an AD73311L to configure it for operation in
mixed mode. It is not intended to be a definitive initialization
sequence, but will show users the typical input/output events
that occur in the programming and Operation Phases1. This
description panel refers to Table XX.
Steps 1–3 detail the transfer of the control words to Control
Register A, which programs the device for Mixed-Mode operation. In Step 1, we have the first output sample event following
device reset. The SDOFS signal is raised which prepares the
DSP Rx register to accept the ADC word from the AD73311L.
The device is configured as nonFSLB, which means that the
DSP has control over what is transmitted to the device and in
this case we will not transmit to the device until the output word
has been received from the AD73311L.
In Step 2 the DSP has now received the ADC word. Typically,
an interrupt will be generated following reception of the output
words by the DSP. The transmit register of the DSP is loaded
with the control word destined for the AD73311L. This generates a transmit frame-sync (TFS) that is input to the SDIFS
input of the AD73311L to indicate the start of transmission.
In Step 3 the device has received a control word that addresses
Control Register A and programs the channels into Mixed
Mode-MM and PGM/DATA set to one. Following Step 3, the
device has been programmed into mixed-mode although none of
the analog sections have been powered up (controlled by Control Register C). Steps 4–6 detail update of Control Register B
in mixed-mode. In Steps 4, 5 the ADC sample, which is invalid
as the ADC section is not yet powered up, is transferred to the
DSP’s Rx section. In the subsequent interrupt service routine
Steps 7–10 are similar to Steps 4–6 except that Control Register
C is programmed to power up all analog sections (ADC, DAC,
Reference = 1.2 V, REFOUT). In Step 10, a DAC word is sent
to the device. As 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.
Steps 7–10 illustrate the implementation of Control Register
update and DAC update in a single sample period. Note that
this combination is not possible in the FSLB configuration2.
Steps 11–15 illustrate a Control Register readback cycle. In Step
13, the device has received a Control Word that addresses Control Register C for readback (Bit 14 of the Control Word = 1).
When the device receives the readback request, the register
contents are loaded to the serial register as shown in Step 14.
SDOFS is raised in the device, which causes the readback word
to be shifted out toward the DSP. In Step 15, the DSP has
received the readback word (note that the address field in the
readback word has been decremented to 111b). Steps 16–18
detail an ADC and DAC update cycle using the nonFSLB configuration. In this case no Control Register update is required.
NOTES
1
This sequence assumes that the DSP SPORT's Rx and Tx interrupts are enabled.
It is important to 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.
2
Mixed mode operation with the FSLB configuration is more restricted in that
only a single word can be sent per sample period.
–30–
REV. A
AD73311L
Table XX. Mixed Mode Operation
Step
DSP Tx
AD73311L
DSP Rx
1
DON’T CARE
xxxxxxxxxxxxxxxx
CRA-CH1
1000101011111001
DON’T CARE
xxxxxxxxxxxxxxxx
OUTPUT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
CRA-CH1
1000000000010011
DON’T CARE
xxxxxxxxxxxxxxxx
OUTPUT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRB-CH1
1000100100001001
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
CRB-CH1
1000000100001011
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH1
1000101011111001
DAC WORD
0111111111111111
DAC WORD
1000000000000000
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH1
1000001011111001
DAC WORD
0111111111111111
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH1
10000010xxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH1
10000010xxxxxxxx
READBACK CH 1
1100001011111001
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
READBACK CH 1
1111001011111001
DON’T CARE
xxxxxxxxxxxxxxxx
DAC WORD CH 1
0111111111111111
DAC WORD CH 1
1000000000000000
ADC RESULT CH1
????????????????
DON’T CARE
xxxxxxxxxxxxxxxx
DAC WORD CH 1
0111111111111111
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
????????????????
DON’T CARE
xxxxxxxxxxxxxxxx
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
REV. A
–31–
AD73311L
APPENDIX C
Configuring a Cascade of Two AD73311Ls to Operate in
Data Mode 1
This section describes the typical sequence of control words that
are required to be sent to a cascade of two AD73311Ls to set
them up for data mode operation. In this sequence Registers B,
C and A are programmed before the device enters data mode.
This description panel refers to Table XXI.
At each sampling event, a pair of SDOFS pulses will be observed
which will cause a pair of control (programming) words to be
sent to the device from the DSP. It is advisable that each pair of
control words should program a single register in each device.
The sequence to be followed is Device 2 followed by Device 1.
In Step 1, we have the first output sample event following device
reset. The SDOFS signal is raised on both devices simultaneously,
which prepares the DSP Rx register to accept the ADC word
from Device 2, while SDOFS from Device 1 becomes an SDIFS
to Device 2. As the SDOFS of Device 2 is coupled to the DSP’s
TFS and RFS, and to the SDIFS of Device 1, this event also
forces a new control word to be output from the DSP Tx register to Device 1.
In Step 2, we observe the status of the devices following the
transmission of the first control word. The DSP has received the
output word from Device 2, while Device 2 has received the
output word from Device 1. Device 1 has received the Control
word destined for Device 2. At this stage, the SDOFS of both
devices are again raised because Device 2 has received Device
1’s output word, and as it is not a valid control word addressed
to Device 2, it is passed on to the DSP. Likewise, Device 1 has
received a control word destined for Device 2-address field is
not zero-and it decrements the address field of the control word
and passes it on.
Step 3 shows completion of the first series of control word
writes. The DSP has now received both output words and each
device 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. Note that both
devices are updated simultaneously as 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 devices can result in corrupted operation. This
will not happen in the case of an FSLB configuration as shown
here, but must be taken into account in a nonFSLB configuration. One other important observation of this sequence is that
the data words are received and transmitted in reverse order,
i.e., the ADC words are received by the DSP, Device 2 first,
then Device 1 and, similarly, the transmit words from the DSP
are sent to Device 2 first, then to Channel 1. This ensures that
all devices are updated at the same time. Steps 4–6 are similar to
Steps 1–3 but, instead, program Control Register C to power-up
the analog sections of the device (ADCs, DACs and reference).
Steps 7–9 are similar to Steps 1–3 but, instead, program Control Register A, with a device count field equal to two devices in
cascade and sets the PGM/DATA bit to one to put the device
in data mode.
In Step 10, the programming phase is complete and we now
begin actual device data read and write. The words loaded into
the serial registers of the two devices at the ADC sampling
event now contain valid ADC data and the words written to the
devices from the DSP’s Tx register will now be interpreted as
DAC words. The DSP Tx register contains the DAC word for
Device 2.
In Step 11, the first DAC word has been transmitted into the
cascade and the ADC word from Device 2 has been read from
the cascade. The DSP Tx register now contains the DAC
word for Device 1. As the words being sent to the cascade are
now being interpreted as 16-bit DAC words, the addressing
scheme now changes from one where the address was embedded in the transmitted word, to one where the serial port now
counts the SDIFS pulses. When the number of SDIFS pulses
received equals the value in the Device count field of Control
Register A, the length of the cascade-each device updates its
DAC register with the present word in its serial register. In
Step 11 each device has received only one SDIFS pulse; Device
2 received one SDIFS from the SDOFS of Device 1 when it
sent its ADC word, and Device 1 received one SDIFS pulse
when it received the DAC word for Device 2 from the DSP’s
Tx register. Therefore, each device raises its SDOFS line to
pass on the current word in its serial register, and each device
now receives another SDIFS pulse.
Step 12 shows the completion of an ADC read and DAC write
cycle. Following Step 11, each device has received two SDIFS
pulses that equal the setting of the device count field in Control
Register A. The DAC register in each device is now updated
with the contents of the word that accompanied the SDIFS
pulse that satisfied the device count requirement. The internal
frame sync counter is now reset to zero and will begin counting
for the next DAC update cycle. Steps 10–12 are repeated on
each sampling event.
NOTE
1
This sequence assumes that the DSP SPORT’s Rx and Tx interrupts are
enabled. It is important to 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.
–32–
REV. A
AD73311L
Table XXI. Data Mode Operation
Step
1
2
3
4
5
6
7
8
9
10
11
12
REV. A
DSP
Tx
AD733111L
Device 1
AD73311L
Device 2
DSP
Rx
CRB-CH2
1000100100001011
CRB-CH1
1000000100001011
CRC-CH2
1000101011111001
OUTPUT CH1
0000000000000000
CRB-CH2
1000100100001011
CRB-CH1
1000000100001011
OUTPUT CH2
0000000000000000
OUTPUT CH1
0000000000000000
CRB-CH2
1000000100001011
DON’T CARE
xxxxxxxxxxxxxxxx
OUTPUT CH2
0000000000000000
OUTPUT CH1
0000000000000000
CRC-CH2
1000101011111001
CRC-CH1
1000001011111001
CRA-CH2
1000100000010001
OUTPUT CH1
1000000100001011
CRC-CH2
1000101011111001
CRC-CH1
1000001011111001
OUTPUT CH2
1000000100001011
OUTPUT CH2
1011100100001011
CRC-CH2
1000001011111001
DON’T CARE
xxxxxxxxxxxxxxxx
OUTPUT CH2
1011100100001011
OUTPUT CH1
1011000100001011
CRA-CH2
1000100000010001
CRA-CH1
1000000000010001
CRB-CH2
0111111111111111
OUTPUT CH1
1000001011111001
CRA-CH2
1000100000010001
CRA-CH1
1000000000010001
OUTPUT CH2
1000001011111001
OUTPUT CH2
1011101011111001
CRA-CH2
1000000000010001
DON’T CARE
xxxxxxxxxxxxxxxx
OUTPUT CH2
1011101011111001
OUTPUT CH1
1011001011111001
DAC WORD CH 2
0111111111111111
DAC WORD CH 1
1000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
????????????????
DAC WORD CH 2
0111111111111111
DAC WORD CH 1
1000000000000000
ADC RESULT CH2
????????????????
ADC RESULT CH1
????????????????
DAC WORD CH 2
0111111111111111
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH2
????????????????
ADC RESULT CH1
????????????????
–33–
AD73311L
APPENDIX D
Configuring a cascade of two AD73311Ls to Operate in Mixed
Mode
This section describes a typical sequence of control words that
would be sent to a cascade of two AD73311Ls to configure
them for operation in mixed mode. It is not intended to be a
definitive initialization sequence, but will show users the typical
input/output events that occur in the programming and operation phases.1 This description panel refers to Table XXII.
Steps 1–5 detail the transfer of the control words to Control
Register A, which programs the devices for Mixed-Mode operation. In Step 1, we have the first output sample event following
device reset. The SDOFS signal is simultaneously raised on both
devices, which prepares the DSP Rx register to accept the ADC
word from Device 2 while SDOFS from Device 1 becomes an
SDIFS to Device 2. The cascade is configured as nonFSLB,
which means that the DSP has control over what is transmitted
to the cascade2 and in this case we will not transmit to the
devices until both output words have been received from the
AD73311Ls.
In Step 2, we observe the status of the devices following the
reception of the Device 2 output word. The DSP has received
the ADC word from Device 2, while Device 2 has received the
output word from Device 1. At this stage, the SDOFS of Device
2 is again raised because Device 2 has received Channel 1’s
output word and, as it is not addressed to Device 2, it is passed
on to the DSP.
In Step 3 the DSP has now received both ADC words. Typically, an interrupt will be 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
Device 2. This generates a transmit frame-sync (TFS) that is
input to the SDIFS input of the AD73311L (Device 1) to indicate the start of transmission.
In Step 4, Device 1 now contains the Control Word destined for
Device 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 Device 1 (this can be done using
autobuffering of transmit or by handling transmit interrupts
following each word sent).
In Step 5 each device has received a control word that addresses
Control Register A and sets the device count field equal to two
devices and programs the devices into Mixed Mode—MM and
PGM/DATA set to one. 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 6–10 detail update of Control Register B in mixedmode. In Steps 6–8 the ADC samples, which are invalid as the
ADC section is not yet powered up, are transferred to the DSP’s
Rx section. In the subsequent interrupt service routine the Tx
Register is loaded with the control word for Device 2. In Steps
9–10, Devices 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 11–17 are similar to Steps 6–12 except that Control Register C is programmed to power up all analog sections (ADC,
DAC, Reference = 2.4 V, REFOUT). In Steps 16–17, DAC
words are sent to the device—both DAC words are necessary
as each device will only update its DAC when the device has
counted a number of SDIFS pulses, accompanied by DAC words
(in mixed-mode, the MSB = 0), that is equal to the device count
field of Control Register A3. As the devices 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 Device 1.
Steps 11–17 illustrate the implementation of Control Register
update and DAC update in a single sample period. Note that
this combination is not possible in the FSLB configuration.2
Steps 18–25 illustrate a Control Register readback cycle. In Step
22, both devices have received a Control Word that addresses
Control Register C for readback (Bit 14 of the Control Word =
1). When the devices receive the readback request, the register
contents are loaded to the serial registers as shown in Step 23.
SDOFS is raised in both devices, which causes these readback
words to be shifted out toward the DSP. In Step 24, the DSP
has received the Device 2 readback word while Device 2 has
received the Device 1 readback word (note that the address field
in both words has been decremented to 111b). In Step 25, the
DSP has received the Device 1 readback word (its address field
has been further decremented to 110b).
Steps 26–30 detail an ADC and DAC update cycle using the
nonFSLB configuration. In this case no Control Register update
is required.
NOTES
1
This sequence assumes that the DSP SPORT’s Rx and Tx interrupts are
enabled. It is important to 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.
2
Mixed mode operation with the FSLB configuration is more restricted in that
the number of words sent to the cascade equals the number of devices in the
cascade, which means that DAC updates may need to be substituted with a
register write or read.
3
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.
–34–
REV. A
AD73311L
Table XXII. Mixed Mode Operation
Step
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
REV. A
DSP
Tx
AD73311L
Device 1
AD73311L
Device 2
DSP
Rx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRA-CH2
1000101011111001
CRA-CH1
1000000000010011
DON’T CARE
xxxxxxxxxxxxxxxx
OUTPUT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRA-CH2
1000100000010011
CRA-CH1
1000000000010011
OUTPUT CH2
0000000000000000
OUTPUT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRA-CH2
1000000000010011
DON’T CARE
xxxxxxxxxxxxxxxx
OUTPUT CH2
0000000000000000
OUTPUT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRB-CH2
1000100100001011
CRB-CH1
1000000100001011
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRB-CH2
1000100100001011
CRB-CH1
1000000100001011
ADC RESULT CH2
0000000000000000
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRB-CH2
1000000100001011
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH2
0000000000000000
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH2
1000101011111001
CRC-CH1
1000001011111001
DAC WORD CH 2
0111111111111111
DAC WORD CH 1
1000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH2
1000101011111001
CRC-CH1
1000001011111001
DAC WORD CH 2
0111111111111111
DAC WORD CH 1
1000000000000000
ADC RESULT CH2
0000000000000000
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH2
1000001011111001
DON’T CARE
xxxxxxxxxxxxxxxx
DAC WORD CH 2
0111111111111111
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH2
0000000000000000
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH2
11001010xxxxxxxx
CRC-CH1
10000010xxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH2
11001010xxxxxxxx
CRC-CH1
10000010xxxxxxxx
READBACK CH 1
1100001011111001
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH2
0000000000000000
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
CRC-CH2
10000010xxxxxxxx
READBACK CH 2
1100001011111001
READBACK CH 1
1111101011111001
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH2
0000000000000000
ADC RESULT CH1
0000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
READBACK CH 2
1111101011111001
READBACK CH 1
1111001011111001
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DAC WORD CH 2
0111111111111111
DAC WORD CH 1
1000000000000000
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH1
????????????????
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DAC WORD CH 2
0111111111111111
DAC WORD CH 1
1000000000000000
ADC RESULT CH2
????????????????
ADC RESULT CH1
????????????????
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
DAC WORD CH 2
0111111111111111
DON’T CARE
xxxxxxxxxxxxxxxx
ADC RESULT CH2
????????????????
ADC RESULT CH1
????????????????
DON’T CARE
xxxxxxxxxxxxxxxx
DON’T CARE
xxxxxxxxxxxxxxxx
–35–
AD73311L
APPENDIX E
DAC Timing Control Example
SE
The AD73311’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 39 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
will have received the 16-bit word. The DSP may process this
information and generate a DAC word to be sent to the AD73311.
Time t3 marks the beginning of the sequence of sending the
DAC word to the AD73311. This sequence ends at time t4
where the DAC register will be updated from the 16 bits in the
AD73311’s serial register. However, the DAC will not be updated
from the DAC register until time t5, which may not be acceptable 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 VIII for details of DAC Timing Control settings).
SCLK
C00689a–2.5–8/00 (rev. A)
SDOFS
ADC
WORD
SDO
SDIFS
DAC
WORD
SDI
DAC
REGISTER
UPDATE
DAC LOAD
FROM DAC
REGISTER
t1
t2
t3
t4
t6
t5
Figure 39. DAC Timing Control
OUTLINE DIMENSIONS
Dimensions shown in inches and (mm).
20-Lead Small Outline IC (R-20)
11
1
10
PIN 1
0.1043 (2.65)
0.0926 (2.35)
0.0291 (0.74)
ⴛ 45ⴗ
0.0098 (0.25)
8ⴗ 0.0500 (1.27)
0.0500 0.0192 (0.49)
0ⴗ 0.0157 (0.40)
(1.27) 0.0138 (0.35) SEATING 0.0125 (0.32)
PLANE
BSC
0.0091 (0.23)
20-Lead Shrink Small Outline IC (RS-20)
20-Lead Thin Shrink Small Outline IC (RU-20)
0.295 (7.50)
0.271 (6.90)
0.260 (6.60)
0.252 (6.40)
20
11
0.311 (7.9)
0.301 (7.64)
1
11
0.177 (4.50)
0.169 (4.30)
0.212 (5.38)
0.205 (5.21)
20
PRINTED IN U.S.A.
0.0118 (0.30)
0.0040 (0.10)
0.4193 (10.65)
0.3937 (10.00)
20
0.2992 (7.60)
0.2914 (7.40)
0.5118 (13.00)
0.4961 (12.60)
0.256 (6.50)
0.246 (6.25)
1
10
10
PIN 1
0.07 (1.78)
0.066 (1.67)
0.078 (1.98) PIN 1
0.068 (1.73)
0.008 (0.203)
0.002 (0.050)
0.006 (0.15)
0.002 (0.05)
0.0256
(0.65)
BSC
SEATING 0.009 (0.229)
PLANE
0.005 (0.127)
8ⴗ
0ⴗ
SEATING
PLANE
0.037 (0.94)
0.022 (0.559)
–36–
0.0433 (1.10)
MAX
0.0256 (0.65) 0.0118 (0.30)
BSC
0.0075 (0.19)
0.0079 (0.20)
0.0035 (0.090)
8ⴗ
0ⴗ
0.028 (0.70)
0.020 (0.50)
REV. A