Advanced Multi-bit Delta Sigma Architecture
24-bit Conversion
Automatic Detection of Sample Rates up to
192 kHz
103 dB Dynamic Range
-88 dB THD+N
Single-Ended Output Architecture
Direct Stream Digital
®
(DSD
–Non-Decimating Volume Control
–On-Chip 50 kHz Filter
–Matched PCM and DSD Analog Ou tput
Levels
Compatible with Industry-Standard Time
Division Multiplexed (TDM) Serial Interface
Selectable Digital Filters
Volume Control with 1/2-dB Step Size and Soft
Ramp
Low Clock-Jitter Sensitivity
+5 V Analog Supply, +2.5 V Digital Supply
Separate 1.8 to 5 V Logic Supplies for the
Control and Serial Ports
™)
Mode
Description
The CS4384 is a complete 8-channel digital-to-analog
system. This D/A system includes digital de-emphasis,
half-dB step size volume control, ATAPI channel mixing, selectable fast and slow digital interpolation filters
followed by an oversampled, multi-bit delta sigma modulator which includes mismatch shaping technology that
eliminates distortion due to capacitor mismatch. Following this stage is a multi-element switched capacitor
stage and low-pass filter with single-ended analog
outputs.
The CS4384 also has a proprietary DSD processor
which allows for volume control and 50 kHz on-chip filtering without an intermediate decimation stage. It also
offers an optional path for direct DSD conve r sion by directly using the multi-element switched capacitor array.
The CS4384 accepts PCM data at sample rates from
4 kHz to 216 kHz, DSD audio data, and delivers excellent sound quality. These featu res are ideal for multichannel audio systems including SACD players, A/V receivers, digital TV’s, mixing consoles, effects
processors, and sound cards.
This product is available in 48-pin LQFP package in
Commercial (-40°C to +85°C) temperature grade. See
“Ordering Information” on page 5 1 for complete details.
Table 10. Example Digital Volume Settings.............................................................................................. 44
CS4384
DS620F15
1. PIN DESCRIPTION
DSD3
DSD2
DSD1
GND
MCLK
LRCK
SDIN1
SCLK
M4(TST)
SDIN2
M3(TST)
DSD6
DSD5
DSD7
SDIN4
CS4384
M1(SDA/CDIN)
M2(SCL/CCLK)
DSD8
M0(AD0/CS)
DSD4
48 47 46 45 44 43 42 41 40 39 38 37
1
2
3
4
VD
5
6
7
8
9
10
11
2
1
13 14 15 16 17 18 19 20 21 22
SDIN3
VLS
VLC
CS4384
AOUT2
AOUT1
23 24
MUTEC2
TST_OUT
TST_OUT
36
35
34
33
32
31
30
29
28
27
26
25
AOUT8
TST_OUT
AOUT3
AOUT4
TST_OUT
VA
GND
TST_OUT
AOUT5
AOUT6
TST_OUT
TST_OUT
AOUT7
TST_OUT
MUTEC1
DSD_SCLK
VQ
RST
FILT+
Pin Name#Pin Description
VD4
GND
MCLK6
LRCK7
SDIN1
SDIN2
SDIN3
SDIN4
SCLK9SerialClock (Input) - Serial clocks for the serial audio interface.
VLC18
RST
FILT+20
VQ21Quiescent Voltage (Output) - Filter connection for internal quiescent voltage.
MUTEC1
MUTEC234
Digital Power (Input) - Positive power supply for the digital section. Refer to the Recommended
Operating Conditions for appropriate voltages.
5
Ground (Input) - Ground reference. Should be connected to analog ground.
31
Master Clock (Input) - Clock source for the delta-sigma modulator and digital filters. Table 1 illus-
trates several standard audio sample rates and the required master clock frequencies.
Left Right Clock (Input) - Determines which channel, Left or Right, is currently active on the serial
audio data line. The frequency of the left/right clock must be at the audio sample rate, Fs.
8
11
Serial Data Input (Input) - Input for two’s complement serial audio data.
13
14
Control Port Power (Input) - Determines the required signal level for the control port and hardware mode configuration pins. Refer to the Recommended Operating Conditions fo r appropriate
voltages.
Reset (Input) - The device enters a low power mode and all internal registers are reset to their
19
default settings when low.
Positive Voltage Reference (Output) - Positive reference voltage for the internal sampling cir-
cuits. Requires the capacitive decoupling to analog ground as shown in the Typical Connection
Diagram.
4122Mute Control (Output) - These pins are intended to be used as a control for external mute circuits
on the line outputs to prevent the clicks and pops that can occur in any single supply system.
6DS620F1
CS4384
Pin Name#Pin Description
AOUT1
AOUT2
AOUT3
AOUT4
AOUT5
AOUT6
AOUT7
AOUT8
VA32
VLS43
TST_OUT
Hardware Mode Definitions
M0
M1
M2
M3
M4
Software Mode Definitions
SCL/CCLK15
SDA/CDIN16
AD0/CS17
TST
DSD Definitions
DSD1, DSD2
DSD3, DSD4
DSD5, DSD6
DSD7, DSD8
DSD_SCLK42DSD Serial Clock(Input) - Serial clock for the Direct Stream Digital serial audio interface.
39
38
35
34
29
28
25
24
23, 26
27, 30
33, 36
37, 40
17
16
15
12
10
10
12
3, 2
1, 48
47,46
45, 44
Analog Output (Output) - The full scale analog output level is specified in the Analog Characteris-
tics specification table.
Analog Power (Input) - Positive power supply for the analog section. Refer to the Recommended
Operating Conditions for appropriate voltages.
Serial Audio Interface Power (Input) - Determines the required signal level for the serial audio
interface. Refer to the Recommended Operating Conditions for appropriate voltages.
Test Output - These pins need to be floating and not connected to any trace or plane.
Mode Selection (Input) - Determines the operational mode of the device as detailed in Tables 4
and 5.
Serial Control Port Clock (Input) - Serial clock for the serial control port. Requires an external
pull-up resistor to the logic interface voltage in I²C
gram.
Serial Control Port Data (Input/Output) - SDA is a data I/O line in I²C mode and is open drain,
requiring an external pull-up resistor to the logic interface voltage, as shown in the Typical Connection Diagram; CDIN is the input data line for the control port interface in SPI
Address Bit 0 (I²C) / Control Port Chip Select (SPI) (Input) - AD0 is a chip address pin in I²C
mode; CS
Test - T hese pins need to be tied to analog ground.
Direct Stream Digital Input (Input) - Input for Direct Stream Digital serial audio data.
is the chip select signal for SPI mode.
®
mode as shown in the Typical Connection Dia-
™
mode.
DS620F17
CS4384
2. CHARACTERISTICS AND SPECIFICATIONS
RECOMMENDED OPERATING CONDITIONS
(GND = 0 V; all voltages with respect to ground.)
ParametersSymbol Min TypMaxUnits
DC Power Supply Analog power
Digital internal power
Serial data port interface power
Control port interface power
Ambient Operating Temperature (Power Applied) -CQZT
VA
VD
VLS
VLC
A
4.75
2.37
1.71
1.71
-40-+85°C
5.0
2.5
5.0
5.0
5.25
2.63
5.25
5.25
V
V
V
V
ABSOLUTE MAXIMUM RATINGS
(GND = 0 V; all voltages with respect to ground.)
ParametersSymbolMinMaxUnits
DC Power Supply Analog power
Digital internal power
Serial data port interface power
Control port interface power
Input Current Any Pin Except SuppliesI
Digital Input Voltage Serial data port interface
Control port interface
Ambient Operating Temperature (power applie d)T
Storage TemperatureT
WARNING:Operation at or beyond these limits may result in permanent damage to the device. Normal operation
is not guaranteed at these extremes.
VD
VLS
VLC
V
IND-S
V
IND-C
VA
in
op
stg
-0.3
-0.3
-0.3
-0.3
-±10mA
-0.3
-0.3
-55125°C
-65150°C
6.0
3.2
6.0
6.0
VLS+ 0.4
VLC+ 0.4
V
V
V
V
V
V
8DS620F1
CS4384
DAC ANALOG CHARACTERISTICS
Test Conditions (unless otherwise indicated): VA = VLS = VLC = 5 V; VD = 2.5 V; TA = 25 °C; Full-Scale 997 Hz
input sine wave
“Typical Connection Diagram” on page 18; Measurement Bandwidth 10 Hz to 20 kHz.
Interchannel Gain Mismatch-0.1-dB
Gain Drift-100-ppm/°
Analog Output
Full Scale Differential- PCM, DSD processor
Output Voltage (Note 3)Direct DSD Mode
Output Impedance Z
Max DC Current draw from an AOUT pinI
Min AC-Load ResistanceR
Max Load CapacitanceC
Quiescent VoltageV
Max Current draw from V
(Note 1); Tested under max ac-load resistance; Valid with FILT+ and VQ capacitors as shown in
ParametersSymbolMinTypMaxUnit
16-bit A-weighted
(Note 2) unweighted
(Note 2) 16-bit 0 dB
Q
unweighted
-20 dB
-60 dB
-20 dB
-60 dB
THD+N-
V
FS
64%•V
47%•V
OUT
OUTmax
L
L
Q
I
QMAX
97
94
-
-
-
-
-
-
-
A
A
-130-Ω
-1.0-mA
-3-kΩ
-100-pF
- 50% VA-VDC
-10-µA
103
100
97
94
-88
-80
-40
-88
-74
-34
66%•V
48%•V
-
-
-
-
-82
-74
-34
-
-
-
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
C
A
A
68%•V
49%•V
Vpp
A
Vpp
A
Notes:
1. One-half LSB of triangular PDF dither is added to data.
2. Performance limited by 16-bit quantization noise.
3. V
is tested under load RL and includes attenuation due to Z
FS
OUT
DS620F19
CS4384
POWER AND THERMAL CHARACTERISTICS
ParametersSymbolMinTypMaxUnits
Power Supplies
Power Supply Current normal operation, VA= 5 V
(Note 4) VD= 2.5 V
(Note 5) Interface current, VLC=5 V
VLS=5 V
(Note 6) power-down state (all supplies)
Power Dissipation (Note 4)VA = 5V, VD = 2.5V
normal operation
(Note 6) power-down
Package Thermal Resistancemulti-layer
dual-layer
Power Supply Rejection Ratio (Note 7) (1 kHz)
(60 Hz)
Notes:
4. Current consumption increases with increasing FS within a given spe ed mode and is signal dependant. Max
values are based on highest FS and highest MCLK.
5. I
6. Power Down Mode is defined as RST
measured with no external loading on the SDA pin.
LC
pin = Low with all clock and data lines held static.
7. Valid with the recommended capacitor values on FILT+ and VQ as shown in Figures 7 and 8.
I
A
I
D
I
LC
I
LS
I
pd
θ
JA
θ
JA
θ
JC
PSRR-
-
-
-
-
-
-
-
-
-
-
-
83
20
2
84
200
465
1
48
65
15
60
40
92
25
-
-
-
520
-
-
-
-
-
-
mA
mA
µA
µA
µA
mW
mW
°C/Watt
°C/Watt
°C/Watt
dB
dB
10DS620F1
CS4384
COMBINED INTERPOLATION & ON-CHIP ANALOG FILTER RESPONSE
The filter characteristics have been normalized to the sample rate (Fs) and can be referenced to the desired sample rate by multiplying the given characteristic by Fs.
(See (Note 12))
Parameter
Combined Digital and On-chip Analog Filter Response - Single-Speed Mode - 48 kHz
Passband (Note 9)to -0.01 dB corner
Frequency Response 10 Hz to 20 kHz-0.01-+0.01dB
Combined Digital and On-chip Analog Filter Response - Double-Speed Mode - 96 kHz
Passband (Note 9)to -0.01 dB corner
Frequency Response 10 Hz to 20 kHz-0.01-+0.01dB
StopBand.583--Fs
StopBand Attenuation(Note 10)80--dB
Group Delay-6.15/Fs-s
Combined Digital and On-chip Analog Filter Response - Quad-Speed Mode - 192 kHz
Passband (Note 9) to -0.01 dB corner
Frequency Response 10 Hz to 20 kHz-0.01-+0.01dB
StopBand.635--Fs
StopBand Attenuation(Note 10)90--dB
Group Delay-7.1/Fs-s
to -3 dB corner
Fs = 48 kHz
to -3 dB corner
to -3 dB corner
0
0
-
-
-
0
0
0
0
Fast Roll-Off
-
-
-
-
-
-
-
-
-
.454
.499
±0.36
±0.21
±0.14
.430
.499
.105
.490
UnitMin TypMax
Fs
Fs
dB
dB
dB
Fs
Fs
Fs
Fs
Notes:
8. Slow Roll-off interpolation filter is only available in Software Mode.
9. Response is clock dependent and will scale with Fs.
10. For Single-Speed Mode, the Measurement Bandwidth is from stopband to 3 Fs.
For Double-Speed Mode, the Measurement Bandwidth is from stopband to 3 Fs.
For Quad-Speed Mode, the Measurement Bandwidth is from stopband to 1.34 Fs.
11. De-emphasis is available only in Single-Speed Mode; Only 44.1 kHz De-emphasis is available in Hardware
Mode.
12. Amplitude vs. Frequency plots of this data are available in the “Filter Response Plots” on page 45.
DS620F111
CS4384
COMBINED INTERPOLATION & ON-CHIP ANALOG FILTER RESPONSE
(CONTINED)
Slow Roll-Off (Note 8)
Parameter
Single-Speed Mode - 48 kHz
Passband (Note 9)to -0.01 dB corner
to -3 dB corner
Frequency Response 10 Hz to 20 kHz-0.01-+0.01dB
StopBand.583--Fs
StopBand Attenuation(Note 10)64--dB
Group Delay -7.8/Fs-s
De-emphasis Error (Note 11)Fs = 32 kHz
(Relative to 1 kHz)Fs = 44.1 kHz
Fs = 48 kHz
0
0
-
-
-
-
-
-
-
-
0.417
0.499
±0.36
±0.21
±0.14
Double-Speed Mode - 96 kHz
Passband (Note 9)to -0.01 dB corner
to -3 dB corner
Frequency Response 10 Hz to 20 kHz-0.01-+0.01dB
StopBand.792--Fs
StopBand Attenuation(Note 10)70--dB
Group Delay-5.4/Fs-s
0
0
-
-
.296
.499
Quad-Speed Mode - 192 kHz
Passband (Note 9) to -0.01 dB corner
to -3 dB corner
Frequency Response 10 Hz to 20 kHz-0.01-+0.01dB
StopBand.868--Fs
StopBand Attenuation(Note 10)75--dB
Group Delay-6.6/Fs-s
0
0
-
-
.104
.481
UnitMinTypMax
Fs
Fs
dB
dB
dB
Fs
Fs
Fs
Fs
DSD COMBINED DIGITAL & ON-CHIP ANALOG FILTER RESPONSE
ParameterMinTypMaxUnit
DSD Processor Mode
Passband (Note 9)to -3 dB corner0-50kHz
Frequency Response 10 Hz to 20 kHz-0.05-+0.05dB
Roll-off27--dB/Oct
Direct DSD Mode
Passband (Note 9) to -0.1 dB corner
to -3 dB corner
Frequency Response 10 Hz to 20 kHz-0.1-0dB
Input Sample Rate - LRCK (Auto detect)Single-Spe ed Mode
Double-Speed Mode
Quad-Speed Mode
F
F
F
Fs
Fs
Fs
s
s
s
LRCK Duty Cycle(Note 16)4555%
SCLK Duty Cycle4555%
SCLK High Timet
SCLK Low Timet
LRCK Edge to SCLK Rising Edget
SCLK Rising Edge to LRCK Falling Edge t
SDIN Setup Time Before SCLK Rising Edget
SDIN Hold Time After SCLK Rising Edget
sckh
sckl
lcks
lckd
ds
dh
1-ms
4
50
100
4
84
170
54
108
216
54
108
216
kHz
kHz
kHz
kHz
kHz
kHz
8-ns
8-ns
5-ns
5-ns
3-ns
5-ns
Notes:
14. After powering up, RST
should be held low until after the power supplies and clocks are settled.
15. See Tables 1 - 3 for suggested MCLK frequencies.
16. Not required for TDM Mode.
LRCK
t
sckh
t
t
ds
dh
MSB
SCLK
SDINx
t
lcks
Figure 1. Serial Audio Interface TimingFigure 2. TDM Serial Audio Interface Timing
(128x Oversampled)
DSD_A / _B valid to DSD_SCLK rising setup timet
DSD_SCLK rising to DSD_A or DSD_B hold timet
DSD clock to data transition (Phase Modulation Mode)t
DSD_SCLK
sclkl
sclkh
sdlrs
sdh
dpm
t
sclkl
160--ns
160--ns
1.024
2.048
20--ns
20--ns
-20-20ns
t
sclkh
-
-
3.2
6.4
MHz
MHz
sdlrstsdh
t
DSDxx
Figure 3. Direct Stream Digital - Serial Audio Input Timing
t
dpm
DSD_SCLK
(128Fs)
DSD_SCLK
(64Fs)
DSDxx
Figure 4. Direct Stream Digital - Serial Audio Input Timing for Phase Modulation Mode
t
dpm
DS620F115
SWITCHING CHARACTERISTICS - CONTROL PORT - I²C FORMAT
(Inputs: Logic 0 = GND, Logic 1 = VLC, CL=30pF)
ParameterSymbolMinMaxUnit
CS4384
SCL Clock Frequencyf
Rising Edge to Startt
RST
Bus Free Time Between Transmissionst
Start Condition Hold Time (prior to first clock pulse)t
Clock Low timet
Clock High Timet
Setup Time for Repeated Start Conditiont
SDA Hold Time from SCL Falling (Note 17)t
SDA Setup time to SCL Risingt
Rise Time of SCL and SDAt
Fall Time SCL and SDAt
Setup Time for Stop Conditiont
Acknowledge Delay from SCL Falling t
Notes:
17. Data must be held for sufficient time to bridge the transition time, t
buf
hdst
low
high
sust
hdd
sud
rc
fc
susp
ack
scl
-100kHz
irs
500-ns
4.7-µs
4.0-µs
4.7-µs
4.0-µs
4.7-µs
0-µs
250-ns
, t
rc
, t
fc
-1µs
-300ns
4.7-µs
3001000ns
, of SCL.
fc
RST
SDA
SCL
t
irs
StopS ta rt
t
buf
hdd
t
high
t
sud
t
t
hdst
low
t
Figure 5. Control Port Timing - I²C Format
Repeated
Start
t
t
sust
hdst
Stop
t
f
t
r
t
susp
16DS620F1
SWITCHING CHARACTERISTICS - CONTROL PORT - SPI FORMAT
(Inputs: Logic 0 = GND, Logic 1 = VLC, CL=30pF)
ParameterSymbolMinMaxUnit
CS4384
CCLK Clock Frequencyf
Rising Edge to CS Fallingt
RST
CCLK Edge to CS
CS
High Time Between Transmissionst
Falling to CCLK Edget
CS
CCLK Low Timet
CCLK High Timet
CDIN to CCLK Rising Setup Timet
CCLK Rising to DATA Hold Time(Note 19)t
Rise Time of CCLK and CDIN (Note 20)t
Fall Time of CCLK and CDIN (Note 20)t
Falling (Note 18)t
sclk
srs
spi
csh
css
scl
sch
dsu
dh
r2
f2
Notes:
18. t
only needed before first falling edge of CS after RST rising edge. t
spi
19. Data must be held for sufficient time to bridge the transition time of CCLK.
The CS4384 serially accepts twos complement formatted PCM data at standard audio sample rates including 48,
44.1 and 32 kHz in SSM, 96, 88.2 a nd 64 kHz in DSM, and 192, 176.4 and 128 kHz in QSM. Audio data is input via
the serial data input pins (SDINx). The Left/Right Clock (LRCK) determines which channel is currently being input
on SDINx, and the Serial Clock (SCLK) clocks audio data into the input data buffer. For more information on serial
audio interfaces see AN282 “The 2-Channel Serial Audio Interface: A Tutoria l”.
The CS4384 can be configured in Hardware Mode by the M0, M1, M2, M3 and M4 pins and in Software Mode
through I²C or SPI.
4.1Master Clock
MCLK/LRCK must be an integer ratio as shown in Tables 1 - 3. The LRCK frequency is equal to Fs, the
frequency at which words for each channel are input to the device. The MCLK-to-LRCK frequency ratio and
speed mode is detected automatically during the initialization sequence by counting the number of MCLK
transitions during a single LRCK period and by detecting the absolute speed of MCLK. Internal dividers are
then set to generate the proper internal clocks. Tables 1 - 3 illustrate several standard audio sample rates
and the required MCLK and LRCK frequencies. Please note there is no required phase relationship, but
MCLK, LRCK and SCLK must be synchronous.
= Denotes clock ratio and sample rate combinations which are NOT supported under auto
speed-mode detection. Please see “Switching Characteristics - PCM” on page 14.
20DS620F1
4.2 Mode Select
In Hardware Mode, operation is determined by the Mode Select pins. The states of these pins are continually scanned for any changes; however, the mode should only be changed while the device is in reset
(RST
pin low) to ensure prope r switching from one mod e to another. These pin s require co nnection to supply or ground as outlined in Figure 8. For M0, M1, and M2, supply is VLC. For M3 and M4, supply is VLS.
Tables 4 - 6 show the decode of these pins.
In Software Mode, the operational mode and data format are set in the FM and DIF registers. See “PCM
Control (Address 03h)” on page 36.
CS4384
M1
(DIF1)
M0
(DIF0)
00
01
10
11
M4M3M2
(DEM)
000
001
010
011
100
101
11Table 6
DESCRIPTIONFORMATFIGURE
Left Justified, up to 24-bit data
2
I
S, up to 24-bit data
Right Justified, 16-bit Data
Right Justified, 24-bit Data
Table 4. PCM Digital Interface Format, Hardware Mode Options
M1M0DESCRIPTION
Single-Speed without De-Emphasis (4 kHz to 50 kHz sample rates)
Single-Speed with 44.1 kHz De-Emphasis; see Figure 20
Double-Speed (50 kHz to 100 kHz sample rates)
Table 4
Quad-Speed (100 kHz to 200 kHz sample rates)
Auto Speed-Mode Detect (32 kHz to 200 kHz sample rates)
Auto Speed-Mode Detect with 44.1 kHz De-Emphasis; see Figure 20
DSD Processor Mode
Table 5. Mode Selection, Hardware Mode Options
09
110
211
312
M2M1M0DESCRIPTION
000
001
010
011
100
101
110
111
Table 6. Direct Stre am Digital (DSD), Hardware Mode Options
64x oversampled DSD data with a 4x MCLK to DSD data rate
64x oversampled DSD data with a 6x MCLK to DSD data rate
64x oversampled DSD data with a 8x MCLK to DSD data rate
64x oversampled DSD data with a 12x MCLK to DSD data rate
128x oversampled DSD data with a 2x MCLK to DSD data rate
128x oversampled DSD data with a 3x MCLK to DSD data rate
128x oversampled DSD data with a 4x MCLK to DSD data rate
128x oversampled DSD data with a 6x MCLK to DSD data rate
DS620F121
4.3Digital Interface Formats
The serial port operates as a slave and supports the I²S, Left-Justified, Right-Justified, One-Line Mode
(OLM) and TDM digital interface formats with varying bit depths from 16 to 32 as shown in Figures 9-19.
Data is clocked into the DAC on the rising edge. OLM and TDM configurations are only supported in Software Mode.
CS4384
LRCK
SCLK
SDINx+3 +2 +1+5 +4
MSBLSBMSBLSB
-1 -2 -3 -4 -5
Left Channel
Figure 9. Format 0 - Left-Justified up to 24-bit Data
LRCK
SCLK
SDINx+3 +2 +1+5 +4
LRCK
SCLK
SDINx
MSB
-2 -3 -4 -5
-1
15 14 13 12 11 10
Left Channel
Left Channel
-2 -3 -4
-1
LSBLSB
MSB
-1
-2 -3 -4
Figure 10. Format 1 - I²S up to 24-bit Data
6543210987
15 14 13 12 11 10
Right Channel
+3 +2 +1+5 +4
Right Channel
+3 +2 +1+5 +4
Right Channel
6543210987
Figure 11. Format 2 - Right-Justified 16-bit Data
Left Channel
65432107
Figure 12. Format 3 - Right-Justified 24-bit Data
Right Channel
23 22 21 20 19 18
65432107
LRCK
SCLK
SDINx
32 clocks
0
23 22 21 20 19 18
32 clocks
LRCK
SCLK
SDINx
10
Left Channel
17 1617 16
19 1819 18
15 14 13 12 11 10
Figure 13. Format 4 - Right-Justified 20-bit Data
32 clocks
6543210987
15 14 13 12 11 10
Right Channel
6543210987
22DS620F1
CS4384
LRCK
SCLK
SDINx
10
4.3.1OLM #1
OLM #1 serial audio interface format operates in Single-, Double-, or Quad-Speed Mode and will slave to
SCLK at 128 Fs. Six channels of MSB first 20-bit PCM data are input on SDIN1. The last two channels
are input on SDIN4.
LRCK
SCLK
SDIN1
SDIN4
DAC_A1
20 clks
DAC_A4
20 clks20 clks
Left Channel
17 1617 16
15 14 13 12 11 10
32 clocks
Figure 14. Format 5 - Right-Justified 18-bit Data
64 clks64 clks
Left ChannelRight Channel
LSBMSB
LSBMSBLSBMSBLSBMSBLSBMSBLSBMSBMSB
DAC_A2DAC_A3
20 clks20 clks20 clks20 clks20 clks
6543210987
DAC_B1
DAC_B4
Right Channel
15 14 13 12 11 10
DAC_B2DAC_B3
6543210987
4.3.2OLM #2
OLM #2 serial audio interface format operates in Single-, Double-, or Quad-Speed Mode and will slave to
SCLK at 256 Fs. Six channels of MSB first 24-bit PCM data are input on SDIN1. The last two channels
are input on SDIN4.
LRCK
SCLK
SDIN1
SDIN4
DAC_A1
24 clks
DAC_A4
24 clks24 clks
Figure 15. Format 8 - One Line Mode 1
128 clks
Left ChannelRight Channel
LSBMSB
DAC_A2DAC_A3DAC_B1DAC_B2DAC_B3
LSBMSBLSBMSBLSBMSBLSBMSBLSBMSBMSB
24 clks24 clks24 clks24 clks24 clks
DAC_B4
Figure 16. Format 9 - One Line Mode 2
128 clks
DS620F123
4.3.3OLM #3
OLM #3 serial audio interface format operates in Single-, Double-, or Quad-Speed Mode and will slave to
SCLK at 256 Fs. Eight channels of MSB first 20-bit PCM data are input on SDIN1.
CS4384
LRCK
SCLK
SDIN1
LSBMSB
DAC_A1
20 clks
4.3.4OLM #4
OLM #4 serial audio interface format operates in Single-, Double-, or Quad-Speed Mode and will slave to
SCLK at 256 Fs. Eight channels of MSB first 24-bit PCM data are input on SDIN1.
LRCK
SCLK
SDIN1
LSBMSB
DAC_A1
24 clks
128 clks
Left Channel
LSBMSBLSBMSB
DAC_A2DAC_A3
20 clks20 clks
Figure 17. Format 10 - One Line Mode 3
128 clks
Left Channel
LSBMSBLSBMSB
DAC_A2DAC_A3
24 clks24 clks
Figure 18. Format 11 - One Line Mode 4
DAC_A4
20 clks
DAC_A4
24 clks
128 clks
Right Channel
LSBMSB
LSBMSB
LSBMSBLSBMSBLSBMSBMSB
DAC_B1DAC_B2DAC_B3
20 clks20 clks20 clks
128 clks
Right Channel
LSBMSBLSBMSBLS BMSBMSB
DAC_B1DAC_B2DAC_B3
24 clks24 clks24 clks
DAC_B4
20 clks
DAC_B4
24 clks
LSBMSB
LSBMSB
4.3.5TDM
The TDM serial audio interface format operates in Single-, Double-, or Quad-Speed Mode and will slave
to SCLK at 256 Fs. Data is received most significant bit first on the first SCLK after an LRCK transition
and is valid on the rising edge of SCLK. LRCK identifies the start of a new frame and is equal to the sample
rate, Fs. LRCK is sampled as valid on the rising SCLK edge preceding the most significant bit of the first
data sample and must be held valid for one SCLK period . Each time slot is 32 bits wide, with the valid data
sample left justified within the time slot with the remaining bits being zero padded.
RCK
CLK
DIN1
LSB
256 clks
LSBMSBLSBMSBLSBMSBLSBMSBLSBMSB
DAC_A1DAC_B1DAC_A3DAC_A2
32 clks32 clks32 clks32 clks32 clks
Data
LSBMSB
zero
DAC_B2
32 clks
LSBMSB
DAC_B3
Figure 19. Format 12 - TDM Mode
DAC_A4
32 clks
LSBMSB
LSBMSB
DAC_B4
32 clks
24DS620F1
4.4Oversampling Modes
The CS4384 operates in one of three oversampling mode s based on the input sample rate. Mo de selection
is determined by the M4, M3 and M2 pins in Hardware Mode or the FM bits in Software Mode. Single-Speed
Mode supports input sample rates up to 50 kHz and uses a 128x oversampling ratio. Double-Speed Mode
supports input sample rates up to 100 kHz and uses an oversampling ratio of 64x. Quad-Speed Mode supports input sample rates up to 200 kHz and uses an oversampling ratio of 32x.
The auto speed-mode detect feature allows for the automatic selection of speed mode based off of the incoming sample rate. This allows the CS4384 to accept a wide range of sample rates with no external intervention necessary. The auto speed-mode detect feature is available in both Hardware and Software Mod e.
4.5Interpolation Filter
To accommodate the increasingly complex requirements of digital audio systems, the CS4384 incorporates
selectable interpolation filters for each mode of operation. A “fast” and a “s low” roll-off filter is available in
each of Single-, Double-, or Quad-Speed Modes. These filters have been designed to accommodate a variety of musical tastes and styles. The FILT_SEL bit is used to select which filter is used (see the “Parameter
Definitions” on page 49 for more details).
When in Hardware Mode, only the “fast” roll-off filter is available.
Filter specifications can be found in Section 2, and filter response plots can be found in Figures 28 to51.
CS4384
4.6De-Emphasis
The CS4384 includes on-chip digital de-emphasis filters. The de-emphasis featur e is included to accommodate older audio recordings that utilize pre-emphasis equalization as a means of noise reduction. Figure 20
shows the de-emphasis curve. The frequency response of the de-emphasis curve will scale proportionally
with changes in sample rate, Fs if the input sample rate does not match the coefficient which has been selected.
In Software Mode the required d e- em p ha sis f i lt er co efficients for 32 kHz, 44.1 kHz, or 48 kHz are select ed
via the de-emphasis control bits.
In Hardware Mode only the 44.1 kHz coefficient is available (enabled through the M2 pin). If the input sample rate is not 44.1 kHz and de-emphasis has been selected th en the corner frequencies of the de-emphasis
filter will be scaled by a factor of the actual Fs over 44,100.
Gain
dB
0dB
-10dB
T1=50 µs
T2 = 15 µs
F1F2
3.183 kHz10.61 kHz
Figure 20. De-Emphasis Curve
Frequency
DS620F125
4.7ATAPI Specification
The CS4384 implements the channel mixing functions of the ATAPI CD-ROM specification. The
ATAPI functions are applied per A-B pair. Refer to Table 9 on page 43 and Figure 21 for additional information.
In Software Mode the DSD/PCM bits (Reg. 02h) are used to configure the device for DSD mode. The
DSD_DIF bits (Reg 04h) then control the expected DSD rate and MCLK ratio.
The DIR_DSD bit (Reg 04h) selects between two proprietary methods for DSD to analog conversion. The
first method uses a decimation free DSD processing technique which allows for features such as matched
PCM level output, DSD volume control, and 50kHz on chip filter. The second method sends the DSD data
directly to the on-chip switched-capacitor filter for conversion (without the above mentioned features).
A Channel
Volume
Control
BChannel
Volume
Control
MUTE
MUTE
AoutAx
AoutBx
The DSD_PM_EN bit (Reg. 04h) selects Pha se Modulation (data plus data inver ted) as the style of data
input. In this mode the DSD_PM_Mode bit selects whether a 128Fs or 64x clock is used for phase modulated 64x data (see Figure 22). Use of Phase Modulation Mode may not directly effect the performance of
the CS4384, but may lower the sensitivity to board level routing of the DSD data signals.
The CS4384 can detect errors in the DSD data which does not comply with the SACD specification. The
STATIC_DSD and INVALID_DSD bits (Reg. 04h) allow the CS4384 to alter the incoming in valid DSD data.
Depending on the error, the data may either be attenuated or replaced with a muted DSD signal (the
MUTEC pins would be set according to the DAMUTE bit (Reg. 08h)).
More information for any of these register bits can be found in the “Parameter Definitions” on page 49.
The DSD input structure and analog outputs are designed to handle a nominal 0 dB-SACD (50% modulation
index) at full rated performance. Signals of +3 dB-SACD may be applied for brief periods of time however,
performance at these levels is not guaranteed. If sustained +3 dB-SACD levels are required, the digital volume control should be set to -3.0 dB. This same volume control register affects PCM output levels. There
is no need to change the volume control setting between PCM and DSD in order to have the 0 dB output
levels match (both 0 dBFS and 0 dB-SACD will output at -3 dB in this case).
26DS620F1
DSD Normal Mode
CS4384
DSD Phase
Modulation Mode
Not Used
BCKA
(64Fs)
DSD_SCLK
Not Used
DSDAx,
DSDBx
D1
D1D0D2
Figure 22. DSD Phase Modulation Mode Diagram
4.9Grounding and Power Supply Arrangements
As with any high resolution converter, the CS4384 requires careful attention to power supply and grounding
arrangements if its potential performance is to be realized. The Typical Connection Diagram shows the recommended power arrangements, with VA, VD, VLC, and VLS connected to clean supplies. If the ground
planes are split between digital ground and analog ground, the GND pins of the CS4384 should be connected to the analog ground plane.
D1
D2D0
DSD_SCLK
DSD_SCLK
DSDAx,
DSDBx
Not Used
BCKA
(128Fs)
BCKD
(64Fs)
All signals, especially clocks, should be kept away from the FILT+ and VQ pins in order to avoid unwanted
coupling into the DAC.
4.9.1Capacitor Placement
Decoupling capacitors should be placed as close to the DAC as possible, with the low value ceram ic capacitor being the closest. To further minimize impedance, these capacitors should be located on the same
layer as the DAC. If desired, all supply pins with similar voltage ratings may be connected to the same
supply, but a decoupling capacitor should still be placed on each supply pin.
Note: All decoupling capacitors should be referenced to analog ground.
The CDB4384 evaluation board demonstrates the optimum layout and power supply arrangements.
4.10Analog Output and Filtering
The CS4384 does not include phase or amplitude compensation for an external filter. Therefore, the DAC
system phase and amplitude response will be dependent on the external analog circuitry.
Figure 23 shows how the full-scale analog output level specification is derived.
Figure 24 shows how the recommended outp ut filtering with location for optional mute circuit.
DS620F127
CS4384
4.175 V
AOUT
4.11The MUTEC Outputs
The MUTEC1 and MUTEC234 pins have an auto-polarity detect feature. The MUTEC output pins are high
impedance at the time of r eset. The exte rnal mute circ uitry needs to be self biased into an active state in
order to be muted during reset. Upon release of reset, the CS438 4 will detect the status of the MUTEC pins
(high or low) and will then select that state as the polarity to drive when the mutes become active. The external-bias voltage level that th e MUTEC pins see a t the time of rele ase of reset must m eet the “MUTEC
auto detect input high/low voltage” specs as outlined in the Digital Characteristics section.
2.5 V
0.825 V
Full-Scale Output Level= AOUT= 3.35 Vpp
Figure 23. Full-Scale Output
Figure 24. Recommended Output Filter
Figure 25 shows a single example of both an a ctive high and an acti ve low mute dr ive circuit. In these de -
signs, the pull-up and pull-down resistors have been especially chosen to meet the input high/low th reshold
when used with the MMUN2111 and MMUN2211 internal bias resistances of 10 kΩ.
Use of the Mute Control function is not mandatory but recommended for designs requiring the absolute minimum in extraneous clicks and pops. Also, use of the Mute Control function can enable the system designer
to achieve idle channel noise/signal-to-noise ratios which are only limited by the external mute circuit.
28DS620F1
Figure 25. Recommended Mute Circuitry
4.12Recommended Power-Up Sequence
CS4384
4.12.1Hardware Mode
1. Hold RST low until the power supplies and configuration pins are stable, and the master and left/right
clocks are locked to the appropriate frequencies, as discussed in Section 4.1. In this state, the
registers are reset to the default settings, FILT+ will remain low, and VQ will be connected to VA/2.
If RST
can not be held low long enough the SDINx pins should remain static low until all other clocks
are stable, and if possible the RST
2. Bring RST
Hardware power-up sequence after approximately 512 LRCK cycles in Single-Speed Mode (1024
LRCK cycles in Double-Speed Mode, and 2048 LRCK cycles in Quad-Speed Mode).
high. The device will remain in a low power state with FILT+ low and will initiate the
4.12.2Software Mode
1. Hold RST low until the power supply is stable, and the master and left/right clocks are loc ked to the
appropriate frequencies, as discussed in Section 4.1. In this state, the register s are reset to the default
settings, FILT+ will remain low, and VQ will be connected to VA/2.
2. Bring RST
Single-Speed Mode (1024 LRCK cycles in Double-Speed Mode, and 2048 LRCK cycles in QuadSpeed Mode).
3. In order to reduce the chances of clicks and pops, perform a write to the CP_EN bit prior to the
completion of approximately 512 LRCK cycles in Single-Speed Mode (1024 LRCK cycles in DoubleSpeed Mode, and 2048 LRCK cycles in Quad-Speed Mode). The desired register settings can be
loaded while keeping the PDN bit set to 1. Set the RMP_UP and RMP_DN bits to 1, then set the
format and mode control bits to the desired settings.
high. The device will remain in a low power state with FILT+ low for 512 LRCK cycles in
should be toggled low again once the system is stable.
If more than the stated number of LRCK cycles passes before CPEN bit is written then the chip will
enter Hardware Mode and begin to operate with the M0-M4 as the mode settings. CPEN bit may be
written at anytime, even after the Hardware sequ e nc e ha s begu n. It is advised tha t if the CPEN bit
can not be set in time then the SDINx pins should remain static low (this way no audio data can be
DS620F129
converted incorrectly by the Hardware Mode settings).
4. Set the PDN bit to 0. This will initiate the power-up sequence, which lasts approximately 50 µs.
4.13Recommended Procedure for Switching Operational Modes
For systems where the absolute minimum in clicks and pops is required, it is reco mmended that the MUTE
bits are set prior to changing significant D AC functions (s uch as chan ging sample ra tes or clock sou rces).
The mute bits may then be released after clocks have settled and the proper modes have been set.
It is required to have the device held in reset if the minimum high/low time specs of MCLK can not be met
during clock source changes.
4.14Control Port Interface
The control port is used to load all the internal register settings in order to operate in Software Mode (see
the “Parameter Definitions” on page 49 ). The operation of the contro l port may be completely asynch ronous
with the audio sample rate. However, to avoid potential interference problems, the control port pins should
remain static if no operation is required.
CS4384
The control port operates in one of two modes:
4.14.1MAP Auto Increment
The device has MAP (memory address pointer) auto increment capability enabled by the INCR bit (also
the MSB) of the MAP. If INCR is set to 0, MAP will stay constant for successive
SPI writes. If INCR is set to 1, MAP will auto increment after each byte is written, allowing block reads or
writes of successive registers.
4.14.2I²C Mode
In the I²C Mode, data is clocked into and out of the bi-directional serial control data line, SDA, by the serial
control port clock, SCL (see Figure 26 for the clock to data relationship). There is no CS
ables the user to alter the chip address (001100[AD0][R/W
quired, before powering up the device. If the device ever detects a high to low transition on the AD0/CS
pin after power-up, SPI Mode will be selected.
4.14.2.1 I²C Write
To write to the device, follow the procedure below while adhering to the control port Switching Specifications in Section 2.
1. Initiate a START condition to the
001100. The seventh bit must match the setting of the AD0 pin, and the eighth must be 0. The eighth
bit of the address byte is the R/W
2. Wait for an acknowledge (ACK) from the part, then write to the memory address pointer, MAP. This
byte points to the register to be written.
3. Wait for an acknowledge (ACK) from the part, then write the desired data to the register pointed to by
the MAP.
4. If the INCR bit (see Section 4.14.1) is set to 1, repeat the previous step until all the desired registers
are written, then initiate a STOP condition to the bus.
5. If the INCR bit is set to 0 and further
a repeated START condition and follow the pro cedure detailed from step 1. If no further writes to other
registers are desired, initiate a STOP condition to the bus.
I²C or SPI.
I²C writes or reads and
pin. Pin AD0 en-
]) and should be tied to VLC or GND as re-
I²C bus followed by the address byte. The upper 6 bits must be
bit.
I²C writes to other registers are desired, it is necessary to initiate
30DS620F1
CS4384
4.14.2.2 I²C Read
To read from the device, follow the procedure below while adhering to the control port Switching Specifications.
1. Initiate a START condition to the
I²C bus followed by the address byte. The upper 6 bits must be
001100. The seventh bit must match the setting of the AD0 pin, and the eighth must be 1. The eighth
bit of the address byte is the R/W
bit.
2. After transmitting an acknowledge (ACK), the device will then trans mit the contents of the register
pointed to by the MAP. The MAP register will contain the address of the last register written to the
MAP, or the default address (see Section 4.14.1) if an
I²C read is the first operation performed on the
device.
3. Once the device has transmitted the contents of the register pointed to by the MAP, issue an ACK.
4. If the INCR bit is set to 1, the device will continue to transmit the contents of successive registers. Continue providing a clock and issue an ACK after each byte un til all the de sired reg ister s are re ad, then
initiate a STOP condition to the bus.
5. If the INCR bit is set to 0 and further
a repeated START condition and follow the procedure detailed from steps 1 and 2 from the
instructions followed by step 1 of the
I²C reads from other registers are desired, it is necessary to initiate
I²C Write
I²C Read section. If no further reads from other registers are de-
sired, initiate a STOP condition to the bus.
Note 1
SDA
SCL
Start
001100
ADDR
AD0
R/W
ACK
DATA
1-8
ACK
DATA
1-8
ACK
Stop
Note: If operation is a write, this byte contains the Memory Address Pointer, MAP.
4.14.3SPI Mode
In SPI Mode, data is clocked into the serial control data line, CDIN, by the serial control port clock, CCLK
(see Figure 27 for the clock to data relationship). There is no AD0 pin. Pin CS
is used to control SPI writes to the control port. When the device detects a high to low transition on the
AD0/CS
pin after power-up, SPI Mode will be selected. All signals are inputs and data is clocked in on the
rising edge of CCLK.
4.14.3.1 SPI Write
To write to the device, follow the procedure below while adhe ring to the contr ol port Switching Spe cifications in Section 2.
1. Bring CS
2. The address byte on the CDIN pin must then be 00110000.
3. Write to the memory address pointer, MAP. This byte points to the register to be written.
4. Write the desired data to the register pointed to by the MAP.
5. If the INCR bit (see Section 4.14.1) is set to 1, repeat the previous step until all the desired registers
are written, then bring CS
Figure 26. Control Port Timing, I²C Mode
is the chip select signal and
low.
high.
DS620F131
CS4384
6. If the INC R bit is set to 0 and furt her SPI writ es to other registers are desired, it is necessary to bring
CS
high, and follow the procedure detailed from step 1. If no further writes to other registers are de-
Note: All registers are read/write in I²C Mode and write only in SPI, unless otherwise noted.
6.1Chip Revision (Address 01h)
76543210
PART4PART3PART2PART1PART0REV2REV1REV0
00000- --
6.1.1Part Number ID (PART) [R ead Only]
00000- CS4384
6.1.2Revision ID (REV) [Read Only]
000 - Revision A0
001 - Revision B0
Function:
This read-only register can be used to identify the model and revision number of the device.
6.2Mode Control 1 (Address 02h)
76543210
CPENFREEZEDSD/PCM
00000001
DAC4_DISDAC3_DISDAC2_DISDAC1_DISPDN
6.2.1Control Port Enable (CPEN)
Default = 0
0 - Disabled
1 - Enabled
Function:
This bit defaults to 0, allowing the device to power-up in Stand-Alone Mode. The Control Port Mode can
be accessed by setting this bit to 1. This will allow the operation of the device to be controlled by the registers and the pin definitions will conform to Control Port Mode. To accomplish a clean power-up, the user
should write this bit within 10 ms following the release of Reset.
6.2.2Freeze Controls (FREEZE)
Default = 0
0 - Disabled
1 - Enabled
Function:
This function allows modifications to be made to the registers without the changes taking effect until the
FREEZE is disabled. To make multiple changes in the Control port registers take effect simultaneously,
enable the FREEZE Bit, make all register changes, then Disable the FREEZE bit.
DS620F135
6.2.3PCM/DSD Selection (DSD/PCM)
Default = 0
0 - PCM
1 - DSD
Function:
This function selects DSD or PCM Mode. The appropriate data and clocks should be present before
changing modes, or else MUTE should be selected.
6.2.4DAC Pair Disable (DACx_DIS)
Default = 0
0 - DAC Pair x Enabled
1 - DAC Pair x Disabled
Function:
When the bit is set, the respective DAC channel pair (AOUTAx and AOUTBx) will remain in a reset state.
It is advised that changes to these bits be made while th e power-d own (PDN) bit is enabled to eliminate
the possibility of audible artifacts.
Note:When the device is configured in TDM Mode by setting DIF[3:0] to 1100 (see Digital Interface
Format (DIF)), this function is not available and these bits must be set to 0 for proper op eration.
CS4384
6.2.5Power Down (PDN)
Default = 1
0 - Disabled
1 - Enabled
Function:
The entire device will enter a low-power state when this function is enabled, and the contents of the control
registers are retained in this mode. The power-down bit defaults to ‘enabled’ on power-up and must be
disabled before normal operation in Control Port Mode can occur.
6.3PCM Control (Address 03h)
76543210
DIF3DIF2DIF1DIF0ReservedReservedFM1FM0
00000011
6.3.1Digital Interface Format (DIF)
Default = 0000 - Format 0 (Left Justified, up to 24-bit data )
Function:
These bits select the interface format fo r the serial audio input. The DSD/PCM
PCM or DSD Mode is selected.
bit determines whether
The required relationship between the Left/Right clock, serial clock and serial data is defined by the Digital
Interface Format and the options are detailed in Figures 9-19.
Note:While in PCM Mode, the DIF bits should only be changed when the power-do wn (PDN) bit is set
to ensure proper switching from one mode to another.
Default = 11
00 - Single-Speed Mode (4 to 50 kHz sample rates)
01 - Double-Speed Mode (50 to 100 kHz sample rates)
10 - Quad-Speed Mode (100 to 200 kHz sample rates)
11 - Auto Speed Mode detect (32 kHz to 200 kHz sample rates)
Left Justified, up to 24-bit data
I2S, up to 24-bit data
Right Justified, 16-bit data
Right Justified, 24-bit data
Right Justified, 20-bit data
Right Justified, 18-bit data
One-line Mode 1, 24-bit Data +SDIN4
One-line Mode 2, 20-bit Data +SDIN4
One-line Mode 3, 24-bit 6-channel
One-line Mode 4, 20-bit 6-channel
TDM
All other combinations are Reserved
09
110
211
312
413
514
815
916
1017
1118
1219
Function:
Selects the required range of input sample rates or Auto Speed Mode.
Default = 000 - Format 0 (64x oversampled DSD data with a 4x MCLK to DSD data rate)
Function:
The relationship between the oversampling ratio of the DSD audio data and the required Master clock to
DSD data rate is defined by the Digital Interface Format pins.
The DSD/PCM
DIF2DIF1DIFODESCRIPTION
00064x oversampled DSD data with a 4x MCLK to DSD data rate
00164x oversampled DSD data with a 6x MCLK to DSD data rate
01064x oversampled DSD data with a 8x MCLK to DSD data rate
01164x oversamp l e d DSD data with a 12x MCLK to DSD data rate
100128x oversampled DSD data with a 2x MCLK to DSD data rate.
101128x oversampled DSD data with a 3x MCLK to DSD data rate.
110128x oversampled DSD data with a 4x MCLK to DSD data rate.
111128x oversampled DSD data with a 6x MCLK to DSD data rate.
bit determines whether PCM or DSD Mode is selected.
T able 8. Digital Interface Formats - DSD Mode
DS620F137
6.4.2Direct DSD Conversion (DIR_DSD)
Function:
When set to 0 (default), DSD input data is sent to the DSD processor for filtering and volume control fu nc-
tions.
When set to 1, DSD input data is sent directly to the switched capacitor DACs for a pure DSD conversion.
In this mode, the full-scale DSD and PCM levels will not be matched (see Section 2), the dynamic range
performance may be reduced, the volume control is inactive, and the 50 kHz low-pass filter is not available
(see Section 2 for filter specifications).
6.4.3Static DSD Detect (STATIC_DSD)
Function:
When set to 1 (default), the DSD processor che cks for 28 consecutive zeroes or ones and, if detected,
sends a mute signal to the DACs. The MUTEC pins will eventually go active according to the DAMUTE
register.
When set to 0, this function is disabled.
6.4.4Invalid DSD Detect (INVALID_DSD)
CS4384
Function:
When set to 1, the DSD processor checks for greater than 24 out of 28 bits of the same value and, if de-
tected, will attenuate the data sent to the DACs. The MUTEC pins go active according to the DAMUTE
register.
When set to 0 (default), this function is disabled.
0 - Two Mute control signals
1 - Single mute control signal on MUTEC1
Function:
Selects how the internal mute signals are routed to the MUTEC1 and MUTEC234 pins. When set to ‘0’,
a logical AND of DAC pair 1 mute control signals are output on MUTEC1 and a logical AND of the mute
control signals of DAC pairs 2, 3, and 4 are output on MUTEC234. When set to ‘1’, a logical AND of all
DAC pair mute control signals is output on the MUTEC1 pin, MUTEC234 will remain static. For more information on the use of the mute control function see the MUTEC1 and MUTEC234 p ins in Section 4.11.
6.7.2Channel A Volume = Channel B Volume (Px_A=B)
Default = 0
0 - Disabled
1 - Enabled
Function:
DS620F139
CS4384
The AOUTAx and AOUTBx volume levels are independently controlled by the A and the B Channel Volume Control Bytes when this function is disabled. The volume on both AOUTAx and AOUTBx are determined by the A Channel Attenuation and Volume Control Bytes (per A-B pair), and the B Channel Bytes
are ignored when this function is enabled.
6.7.3Single Volume Control (SNGLVOL)
Default = 0
0 - Disabled
1 - Enabled
Function:
The individual channel volume levels are independently controlled by their respective Volume Control
Bytes when this function is disabled. The volume on all channels is determined by the A1 Channel Volume
Control Byte, and the other Volume Control Bytes are ignored when this function is enabled.
6.8Ramp and Mute (Address 08h)
76543210
SZC1SZC0RMP_UPRMP_DNPAMUTEDAMUTEMUTE_P1MUTE_P0
10111100
6.8.1Soft Ramp and Zero Cross Control (SZC)
Default = 10
00 - Immediate Change
01 - Zero Cross
10 - Soft Ramp
11 - Soft Ramp on Zero Crossings
Function:
Immediate
When Immediate Change is selected all level changes will take effect immediately in one step.
Zero Cross
Zero Cross Enable dictates that signal level changes, either by attenuation changes or muting, will occur
on a signal zero crossing to minimize audible artifacts. The requested level change will occur after a timeout period between 512 and 1024 sample periods (10.7 ms to 21.3 ms at 48 kHz sample rate) if the signal
does not encounter a zero crossing. The zero cross function is independently monitored and impl emented
for each channel.
Soft Ramp
Soft Ramp allows level changes, both muting and attenuation, to be implemented by incrementally ramp-
ing, in 1/8 dB steps, from the current level to the new level at a rate of 1 dB per 8 left/right clock periods.
Soft Ramp on Zero Crossing
Soft Ramp and Zero Cross Enable dictates that signal level changes, either by attenuation changes or
muting, will occur in 1/8 dB steps and be implemented on a signal zero crossing. The 1/8 dB level change
will occur after a time-out period between 512 and 1024 sample periods (10.7 ms to 21.3 ms at 48 kHz
sample rate) if the signal does not encounter a zero crossing. The zero cross function is independently
monitored and implemented for each channel.
Change
40DS620F1
6.8.2Soft Volume Ramp-Up After Error (RMP_UP)
Function:
An un-mute will be performed after executing an LRCK/MCLK ratio change or error, and after changing
the Functional Mode.
When set to 1 (default), this un-mute is effected, similar to attenuation changes, by the Soft and Zero
Cross bits in the Volume and Mixing Control register.
When set to 0, an immediate un-mute is performed in these instances.
Note:For best results, it is recommended that this feature be used in conjunction with the RMP_DN bit.
6.8.3Soft Ramp-Down Before Filter Mode Change (RMP_DN)
Function:
If either the FILT_SEL or DEM bits are changed the DAC will stop conversion for a period of time to
change its filter values. This bit selects how the data is effected prior to and after the change of the filter
values.
When set to 1 (default), a mute will be performed prior to executing a filter mode change and an un-mute
will be performed after executing the filter mode change. This mute and un-mute are effected, similar to
attenuation changes, by the Soft and Zero Cross bits in the Volume and Mixing Control register.
CS4384
When set to 0, an immediate mute is performed prior to executing a filter mode change.
Note:For best results, it is recommended that this feature be used in conjunction with the RMP_UP bit.
6.8.4PCM Auto-Mute (PAMUTE)
Function:
When set to 1 (default) the Digital-to-Analog converter output will mute following the reception of 8192
consecutive audio samples of static 0 or -1. A single sample of non-static data will release the mute. Detection and muting is done independently for each channel. The quiescent voltage on the output will be
retained and the Mute Control pin will go active during the mute period.
When set to 0 this function is disabled.
6.8.5DSD Auto-Mute (DAMUTE)
Function:
When set to 1 (default) the Digital-to-Analog converter output will mute following the reception of 256 re-
peated 8-bit DSD mute patterns (as defined in the SACD specification).
A single bit not fitting the repeated mute pattern (mentioned above) will release the mute. Detection and
muting is done independently for each channel. The quiescent voltage on the output will be retained and
the Mute Control pin will go active during the mute period.
6.8.6MUTE Polarity and DETECT (MUTEP1:0)
Default = 00
00 - Auto polarity detect, selected from MUTEC1 pin
01 - Reserved
10 - Active low mute polarity
11 - Active high mute polarity
DS620F141
Function:
Auto mute polarity detect (00)
See Section 4.11 on page 28 for the description.
Active low mute polarity (10)
CS4384
When RST
released and after this bit is set, the MUTEC output pins will be active low polarity.
Active high mute polarity (11)
At reset time the outputs are high impedance and will need to be biased active. Once reset has been re-
leased and after this bit is set, the MUTEC output pins will be active high polarity.
is low the outputs are high impedance and will need to be biased active. Once reset has been
Function:
The Digital-to-Analog converter output will mute when enabled. The quiescent voltage on the output will
be retained. The muting function is affected, similarly to attenuation changes, by the Soft and Zero Cross
bits. The MUTE pins will go active during the mute period according to the MUTEC bit.
Function:
Selects the appropriate digital filte r to maintain the standard 15 µs/50 µs digital de-emphasis filter re-
sponse at 32, 44.1 or 48 kHz sample rates. (Figure 20 on page 25)
De-emphasis is only available in Single-Speed Mode.
42DS620F1
CS4384
6.10.2ATAPI Channel Mixing and Muting (ATAPI)
Default = 01001 - AOUTAx=aL, AOUTBx=bR (Stereo)
Function:
The CS4384 implements the channel mixing functions of the ATAPI CD-ROM specification. The ATAPI
functions are applied per A-B pair. Refer to Table 9 and Figure 21 for additional information.
These eight registers provide individual volume and mute control for each of the eight channels.
The values for “xx” in the bit fields above are as follows:
Register address 0Bh - xx = A1
Register address 0Ch - xx = B1
Register address 0Eh - xx = A2
Register address 0Fh - xx = B2
Register address 11h - xx = A3
Register address 12h - xx = B3
Register address 14h - xx = A4
Register address 15h - xx = B4
6.11.1Digital Volume Control (xx_VOL7:0)
Default = 00h (0 dB)
Function:
The Digital Volume Control registers allow independent control of the signal levels in 1/2 dB increments
from 0 to -127.5 dB. Volume settings are de coded as shown in Table 10. The volume changes are implemented as dictated by the Soft and Zero Cross bits in the Power and Muting Control register. Note that
the values in the volume setting column in Table 10 are approximate. The actual attenuation is determined
by taking the decimal value of the volume register and multiplying by 6.02/12.
1. How to Achieve Optimum Performance from Delta-Sigma A/D & D/A Converters, by Steven Harris. Paper
presented at the 93rd Convention of the Audio Engineering Society, October 1992.
2. CDB4364 data sheet, available at http://www.cirrus.com.
3. Design Notes for a 2-Pole Filter with Differential Input, by Steven Green. Cirrus Logic Application Note AN48
4. The
5. AN282 “The 2-Channel Serial Audio Interface: A Tutorial”
I²C-Bus Specification: Version 2.0, Philips Semiconductors, December 1998.
http://www.semiconductors.philips.com
9. PARAMETER DEFINITIONS
Total Harmonic Distortion + Noise (THD+N)
The ratio of the rms value of the signal to the rms sum of all other spectral components over the spec ified
bandwidth (typically 10 Hz to 20 kHz), including distortion components. Expressed in decibels.
Dynamic Range
The ratio of the full-scale rms value of the signal to the rms sum of all other spectral components over the
specified bandwidth. Dynamic range is a signal-to-noise measurement over the specified bandwidth made
with a -60 dBFS signal. 60 dB is then added to the resulting measurement to refer the measurement to full
scale. This technique ensures that the distortion components are below the noise le vel and do not affect the
measurement. This measurement technique has bee n accepted by the Audio En gineering Society, AES171991, and the Electronic Industries Association of Japan, EIAJ CP-307.
Interchannel Isolation
A measure of crosstalk between the left and right channels. Measured for each channel at the conver ter's
output with all zeros to the input under test and a full- scale signal applied to the other channel. Units in decibels.
Interchannel Gain Mismatch
The gain difference between left and right channels. Units in decibels.
Gain Error
The deviation from the nominal full-scale analog output for a full-scale digital input.
Gain Drift
The change in gain value with temperature. Units in ppm/°C.
DS620F149
10.PACKAGE DIMENSIONS
48L LQFP PACKAGE DRAWING
D1
D
CS4384
E
E1
1
e
∝
B
A
A1
L
INCHESMILLIMETERS
DIMMINNOMMAXMINNOMMAX
A---0.0550.063---1.401.60
A10.0020.0040.0060.050.100.15
B0.0070.0090.0110.170.220.27
D0.3430.3540.3668.709.0 BSC9.30
D10.2720.280.2806.907.0 BSC7.10
E0.3430.3540.3668.709.0 BSC9.30
E10.2720.280.2806.907.0 BSC7.10
e*0.0160.0200 .0 240.400.50 BSC0.60
L0.0180.240.0300.450.600.75
∝
0.000°4°7.000°0.00°4°7.00°
* Nominal pin pitch is 0.50 mm
Controlling dimension is mm.
JEDEC Designation: MS022
50DS620F1
CS4384
11.ORDERING INFORMATION
ProductDescriptionPackagePb-FreeGradeTemp RangeContainer Order #
CS4384
114 dB, 192 kHz 8-
channel D/A Converter
48-pin
LQFP
YESCommercial -40° to +85° C
CDB4384 CS4384 Evaluation Board ----CDB4384
TrayCS4384-CQZ
Tape & Reel CS4384-CQZR
12.REVISION HISTORY
ReleaseChanges
A1Initial Release
Corrected DAC Pair Disable register de scri p ti o n in “DAC Pair Disable (DACx_DIS)” on page 36
Added note to Digital Interface Format register description in “Digital Interface Format (DIF)” on page 36
Removed Automotive Grade
Added PCM mode format changeable in reset only to “Mode Select” on page 21
F1
Updated ambient operating temperature range for commercial grade
Updated Full Scale Differential Output Voltage in “DAC Analog Characteristics” on page 9
Updated VD power supply current and package thermal resistance in Power and Thermal Characteristics
Updated “Digital Characteristics” on page 13
Updated Legal Information under “IMPORTANT NOTICE” on page 51
Contacting Cirrus Logic Support
For all product questions and inquiries, contact a Cirrus Logic Sales Representative.
To find one nearest you, go to www.cirrus.com.
IMPORTANT NOTICE
Cirrus Logic, Inc. and its subsidiaries ("Cirrus") believe that the informati on contained in this document is accurate and reliable. However, the information is su bject
to change without notice and is provided "AS IS" wit hout warranty of any kind (express or implied). Customers are advised to obtain the latest version of relevant
information to verify, before placing orders, that information being relied on is current and complete. All products are sold subject to the terms and conditions of sale
supplied at the time of order acknowledgm ent, including thos e pertaining to warranty, ind emnification, and li mitation of liability. No responsibility is assumed by Cirrus
for the use of this information, including use of this information as the basis for manufacture or sale of any items, or for infringement of patents or other rights of third
parties. This document is the property of Cirrus and by furnishing this inform atio n, Cirrus gran ts no license , express or implied un der an y patents, mask wor k rights,
copyrights, trademarks, trade secrets or other intellectual property rights. Cirrus owns the copyrights associated with the information contained herein and gives consent for copies to be made of the information only for use with in your organization with respect to Cirrus integrated circuits or other products of Cirrus. This consent
does not extend to other copying such as copying for general distribution, advertising or promotional purposes, or for creating any work for resale.
CERTAIN APPLICATIONS USING SEMI CO NDUCT OR PRODUCT S MAY INVO LVE POT ENTI AL RI SKS OF DEATH, PERS ONAL IN JU RY, OR SEVE RE PROPERTY OR ENVIRONMENTAL DAMAGE (“CRITICAL APPLICATIONS”). CIRRUS PRODUCTS ARE NOT DESIGNED, AUTHORIZED OR WARRANTED FOR USE
IN PRODUCTS SURGICALLY IMPLANTED INTO THE BODY, AUTOMOTIVE SAFETY OR SECURITY DEVICES, LIFE SUPPORT PRODUCTS OR OTHER CRITICAL APPLICATIONS. INCLUSION OF CIRRUS PRODUCTS IN SUCH APPLICATIONS IS UNDERSTOOD TO BE FULLY AT THE CUSTOMER’S RISK AND
CIRRUS DISCLAIMS AND MAKES NO WARRANTY, EXPRESS, STATUTORY OR IMPLIED, INCLUDING THE IMPLIED WARRANTIES OF MERCHANTABILITY
AND FITNESS FOR PARTICULAR PURPOSE, WITH REGARD TO ANY CIRRUS PRODUCT THAT IS USED IN SUCH A MANNER. IF THE CUSTOMER OR
CUSTOMER’S CUSTOMER USES OR PERMITS THE USE OF CIRRUS PRODUCTS IN CRITICAL APPLICATIONS, CUSTOMER AGREES, BY SUCH USE, TO
FULLY INDEMNIFY CIRRUS, ITS OFFICERS, DIRECTORS, EMPLOYEES, DISTRIBUTORS AND OTHER AGENTS FROM ANY AND ALL LIABILITY, INCLUDING ATTORNEYS’ FEES AND COSTS, THAT MAY RESULT FROM OR ARISE IN CONNECTION WITH THESE USES.
Cirrus Logic, Cirrus, and the Cirrus Logic logo designs a re trade ma rks of Cirrus Lo gic, Inc. All othe r brand an d p roduc t names in this docum en t may be trad em arks
or service marks of their respective owners.
I²C is a registered trademark of Philips Semiconductor.
SPI is a trademark of Motorola, Inc.
Direct Stream Digital is a registered trademark of Sony Kabushiki Kaisha TA Sony Corporation.
DSD is a trademark of Sony Kabushiki Kaish a TA So ny Co rporation.
DS620F151
CS4384
52DS620F1
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.