15DEFINITIONS
16LIFE SUPPORT APPLICATIONS
17PURCHASE OF PHILIPS I2C COMPONENTS
2
C-BUS CONTROL
2
S-BUS DESCRIPTION
1998 Apr 272
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
1FEATURES
• SIF input switch e.g. to select between terrestrial TV SIF
and SAT SIF sources
• SIF AGC with 21 dB control range
• SIF 8-bit Analog-to-Digital Converter (ADC)
• DQPSK demodulation for different standards,
simultaneously with 1-channel FM demodulation
• NICAM decoding (B/G, I and L standard)
• Two-carrier multi-standard FM demodulation
(B/G, D/K and M standard)
• Decoding for three analog multi-channel systems
(A2, A2+ and A2*) and satellite sound
• Adaptive de-emphasis for satellite
• Programmable identification (B/G, D/K and M standard)
and different identification times
• Optional AM demodulation for system L, simultaneously
with NICAM
• Monitor selection for FM/AM demodulator outputs and
FM and NICAM signals
• Digital crossbar switch
2
S serial audio output with matrix, level adjust and mute
• I
• Dual audio Digital-to-Analog Converter (DAC) from
digital crossbar switch to analog crossbar switch,
bandwidth 15 kHz
• Analog crossbar switch with inputs for mono and stereo
• Output selection of mono, stereo, dual, dual A or dual B
• 20 kHz bandwidth for analog path
• Standby mode.
2.1Supported standards
The multi-standard/multi-stereo capability of the
TDA9874H is mainly of interest in Europe, but also in
Hong Kong/PR China and South East Asia. This includes
B/G, D/K, I, M and L standard. In other application areas
there exist subsets of those standard combinations or only
single standards are transmitted.
Standard M is transmitted in Europe by the American
Forces Network with European channel spacing
(7 MHz VHF, 8 MHz UHF) and monaural sound.
The AM sound of L/L’ standard is normally demodulated in
the 1st sound IF. The resulting AF signal has to be entered
into the mono audio input of the TDA9874H. A second
possibility is to use the internal AM demodulator stage,
giving limited performance.
Korea has a stereo sound system similar to Europe and is
supported by the TDA9874H. Differences include
deviation, modulation contents and identification. It is
based on M standard.
An overview of the supported standards and sound
systems and their key parameters is given in Table 1.
The analog multi-channel systems are sometimes also
named 2CS (2 carrier systems).
2GENERAL DESCRIPTION
The TDA9874H is a single-chip Digital TV Sound
Demodulator/Decoder (DTVSD1) for analog and digital
multi-channel sound systems in TV/VCR sets and satellite
receivers.
1. For other satellite systems, frequencies of e.g. 5.80 MHz, 6.60 MHz or 6.65 MHz can also be received.
A de-emphasis of 60 µs or in accordance with J17 is available.
2. m/st/d = mono or stereo or dual language sound.
3. Adaptive de-emphasis = compatible to transmitter specification.
OUTL1analog output left
OUTR2analog output right
V
DDA1
V
SSA1
V
SSD1
V
DDD1
V
SSD2
V
DDD2
TP29additional test pin 2; connected to V
NICAM10serial NICAM data output at 728 kHz
TP111additional test pin 1; connected to V
PCLK12NICAM clock output at 728 kHz
ADDR113first I
V
tune
XTALI15crystal oscillator input
XTALO16crystal oscillator output
TEST217test pin 2; connected to V
I
ref
ADDR219second I
V
SSA2
V
DDA2
TEST122test pin 1; connected to V
SIF223sound IF input 2
V
ref1
SIF125sound IF input 1
CRESET26capacitor for power-on reset
V
SSD3
V
DDD3
SCL29I
SDA30I
SDO31I
3analog supply voltage 1; DAC circuitry
4analog ground supply 1; DAC circuitry
5digital ground supply 1; DAC circuitry
6digital supply voltage 1; DAC circuitry
7digital ground supply 2; DSP part
8digital supply voltage 2; DSP part
for normal operation
SSD
for normal operation
SSD
2
C-bus slave address modifier
14tuning voltage output for crystal oscillator
for normal operation
SSD
18resistor for reference current generation; front end circuitry
2
C-bus slave address modifier
20analog ground supply 2; analog front end circuitry
21analog supply voltage 2; analog front end circuitry
for normal operation
SSD
24reference voltage; analog front end circuitry
27digital ground supply 3; front end circuitry
28digital supply voltage 3; front end circuitry
2
C-bus clock input
2
C-bus data input/output
2
S-bus serial data output
1998 Apr 278
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
SYMBOLPINDESCRIPTION
WS32I2S-bus word select input/output
SCK33I
SYSCLK34system clock output
V
V
6.1Description of the demodulator and decoder
section
6.1.1SIF
INPUT
Two input pins are provided. SIF1 e.g. for terrestrial TV
and SIF2 e.g. for a satellite tuner. As no specific filters are
integrated, both inputs have the same specification giving
flexibility in application. The selected signal is passed
through an AGC and then digitized by an 8-bit ADC
running at 24.576 MHz.
6.1.2AGC
The gain of the AGC amplifier is controlled from the ADC
output by means of a digital control loop employing
hysteresis. The AGC has a fast attack behaviour to
prevent ADC overloads, and a slow decay behaviour to
prevent AGC oscillations. For AM demodulation the AGC
must be switched off. When switched off, the control loop
is reset and fixed gain settings can be chosen from a table.
The AGC can be controlled via the I
2
C-bus. Details can be
found in Sections 7.3.1, 7.3.2 and 7.4.6.
6.1.3M
IXER
The digitized input signal is passed on to the mixers, which
mix one or both input sound carriers down to zero IF.
A 24-bit control word for each carrier sets the required
frequency. Access to the mixer control word registers is via
the I2C-bus (see Sections 7.3.4 and 7.3.5). When
receiving NICAM programs, a feedback signal is added to
the control word of the second carrier mixer to establish a
carrier-frequency loop.
6.1.4FM
AND AM DEMODULATION
An FM or AM input signal is passed through a band-limiting
filter onto a demodulator that can be used for either FM or
AM demodulation. Apart from the standard (fixed)
de-emphasis characteristic, an adaptive de-emphasis is
available for Wegener-Panda 1 encoded satellite
programs.
6.1.5FM
DECODING
A two-carrier stereo decoder recovers the left and right
signal channels from the demodulated sound carriers.
Both the European and Korean stereo systems are
supported.
6.1.6FM
IDENTIFICATION
The identification of the FM sound mode is performed by
AM synchronous demodulation of the pilot and
narrow-band detection of the identification frequencies.
The result is available via the I2C-bus interface. A selection
can be made via the I2C-bus for B/G, D/K and M standard,
and for three different time constants that represent
different trade-offs between speed and reliability of
identification.
6.1.7NICAM
DEMODULATION
The NICAM signal is transmitted in a DQPSK code at a bit
rate of 728 kbits/s. The NICAM demodulator performs
DQPSK demodulation and passes the resulting bitstream
and clock signal to the NICAM decoder and, for evaluation
purposes, to pins.
A timing loop controls the frequency of the crystal oscillator
to lock the sampling instants to the symbol timing of the
NICAM data. The polarity of the control signal is selectable
to support applications, in which external circuitry is used
to boost the tuning voltage of the oscillator.
6.1.8NICAM
DECODING
The device performs all decoding functions in accordance
with the
“EBU NICAM 728 specification”
. After locking to
the frame alignment word, the data are descrambled by
application of the defined pseudo-random binary
sequence, and the device synchronizes to the periodic
frame flag bit C0.
The status of the NICAM decoder can be read-out from the
NICAM Status Register by the user (see Section 7.4.2).
The OSB bit indicates that the decoder has locked to the
NICAM data. The VDSP bit indicates that the decoder has
locked to the NICAM data and that the data is valid sound
data. The C4 bit indicates that the sound conveyed by the
FM mono channel is identical to the sound conveyed by
the NICAM channel. The error byte contains the number of
sound sample errors, resulting from parity checking, that
occurred in the past 128 ms period. The Bit Error Rate
(BER) is approximately 0.0000174 times the contents of
the error byte.
BER
bit errors
----------------------total bits
error byte 1.74×10
5–
×≈=
1998 Apr 2710
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
6.1.9NICAM AUTO-MUTE
This function is enabled by setting bit AMUTE LOW
(see Section 7.3.11). Upper and lower error limits may be
defined by writing appropriate values to two registers in the
I2C-bus section (see Sections 7.3.13 and 7.3.14). When
the number of errors in a 128 ms period exceeds the upper
error limit, the auto-mute function will switch the output
sound from NICAM to whatever sound is on the first sound
carrier (FM or AM) or to the analog mono input. When the
error count is smaller than the lower error limit, the NICAM
sound is restored.
The auto-mute function can be disabled by setting bit
AMUTE HIGH. In this case clicks become audible, when
the error count increases. The user will hear a signal of
degrading quality.
A decision to enable/disable the auto-muting is taken by
the microprocessor based on an interpretation of the
application control bits C1, C2, C3 and C4, and possibly
any additional strategy implemented by the setmaker in
the microcontroller software.
When the AM sound in NICAM L systems is demodulated
in the 1st sound IF and the audio signal connected to the
mono input of the TDA9874H, the controlling
microprocessor has to take care of switching from NICAM
reception to mono input, if auto-muting is desired. This
could be achieved by setting the AMSEL bit HIGH
additionally to AMUTE bit LOW (see also Section 7.3.11).
6.1.10C
A circuit diagram of the external components of the
voltage-controlled crystal oscillator is shown in Fig.7 in
Chapter 9.
6.1.11T
All test pins are active HIGH. In normal operation of the
device they can be left open-circuit, as they have internal
pull-down resistors. Test functions are for manufacturing
tests only and are not available to customers.
RYSTAL OSCILLATOR
EST PINS
1998 Apr 2711
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
6.2Description of the DSP
6.2.1L
EVEL SCALING
All input channels to the digital crossbar switch are
equipped with a level adjust facility to change the signal
level in a range of ±15 dB. Adjusting the signal level is
intended to compensate for the different modulation
parameters of the various TV standards. It is
recommended to scale all input channels to be 15 dB
below full-scale (−15 dB (FS)) under nominal conditions.
This will create sufficient headroom to cope with
overmodulation and avoids changes of the volume
impression when switching from FM to NICAM or vice
versa.
6.2.2NICAM
PATH
The NICAM path has a switchable J17 de-emphasis.
6.2.3NICAM
AUTO-MUTE
If NICAM is received, the AUTO-MUTE is enabled and the
signal quality becomes poor, the digital crossbar switches
automatically to FM, Channel 1 or the analog mono input,
as selected by bit AMSEL. This automatic switching
depends on the NICAM bit error rate. The auto-mute
function can be disabled via the I2C-bus.
6.2.5FM
MONITOR
This function provides data words from the
FM demodulator outputs and FM and NICAM signals for
external use, like carrier search or fine tuning. Source
selection and data read-out are performed via the I2C-bus.
6.2.6D
IGITAL CROSSBAR SWITCH
Input channels come from the FM and NICAM paths, while
output channels comprise I2S and the audio DACs to the
analog crossbar switch. Note that there is no connection
from the external analog audio inputs to the digital
crossbar switch.
6.2.7D
IGITAL AUDIO OUTPUT
The digital audio output interface comprises an I2S output
port and a system clock output. The I2S port is equipped
with a level adjust facility that can change the signal level
in a ±15 dB range in 1 dB steps. Muting is possible, too,
and outputs can be disabled to improve EMC
performance.
The I2S-bus output matrix provides the functions of forced
mono, stereo, channel swap, Channel 1 or Channel 2.
6.2.8C
HANNEL TO THE ANALOG CROSSBAR PATH
6.2.4FM (AM)
PATH
A high-pass filter suppresses DC offsets from the
FM demodulator that may occur due to carrier frequency
offsets and supplies the FM monitor function with DC
values, e.g. for the purpose of microprocessor controlled
carrier search or fine-tuning functions.
An adaptive de-emphasis is available for
Wegener-Panda 1 encoded satellite programs.
The de-emphasis stage offers a choice of settings for the
supported TV standards.
The 2 channel decoder performs the dematrixing of
1
⁄2(L + R) and R to L and R signals, of1⁄2(L + R) and
1
⁄2(L − R) to L and R signals or of Channel 1 and
Channel 2 to L and R signals, as demanded by the
different TV standards or user preferences.
A level adjust function is provided with control positions
0 dB, +3 dB, +6 dB and +9 dB in combination with the
audio DACs.
6.2.9G
ENERAL
The level adjust functions can provide signal gain at
multiple locations. Great care has to be taken when using
gain with large input signals, e.g., due to overmodulation,
in order not to exceed the maximum possible signal swing,
which would cause severe signal distortion. The nominal
signal level of the various signal sources to the digital
crossbar switch should be 15 dB below digital full-scale,
i.e., −15 dB (FS).
1998 Apr 2712
This text is here in white to force landscape pages to be rotated correctly when browsing through the pdf in the Acrobat reader.This text is here in
_white to force landscape pages to be rotated correctly when browsing through the pdf in the Acrobat reader.This text is here inThis text is here in
white to force landscape pages to be rotated correctly when browsing through the pdf in the Acrobat reader. white to force landscape pages to be ...
1998 Apr 2713
handbook, full pagewidth
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
NICAM
FM
DC FILTER
FIXED
DE-EMPHASIS
ADAPTIVE
DE-EMPHASIS
FIXED
DE-EMPHASIS
2 CHANNEL
DECODER
LEVEL
ADJUST
LEVEL
ADJUST
DIGITAL
CROSSBAR
SELECT
MATRIX
FM MONITOR
LEVEL
ADJUST
LEVEL
ADJUST
DAC
I
I
MGK755
2
S
2
C
Fig.3 DSP data flow diagram.
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
6.3Description of the analog audio section
6.3.1A
NALOG CROSSBAR SWITCH AND ANALOG MATRIX
The TDA9874H has one external analog stereo input, one
mono input and one two-channel output port. Analog
source selector switches are employed to provide the
desired analog signal routing capability, which is done by
the analog crossbar switch section.
The basic signal routing philosophy of the TDA9874H is
that each switch handles two signal channels at the same
time, e.g. Left and Right, language A and B, directly at the
source. For an overview of the signal flow see Fig.5.
Each source selector switch is followed by an analog
matrix to perform further selection tasks, like putting a
signal from one input channel, say, language A, to both
output channels or for swapping left and right channel.
The analog matrix provides the functions given in Table 6.
All switches and matrices are controlled via the I2C-bus.
6.3.2EXTERNAL AND MONO INPUTS
The external and mono inputs accept signal levels of up to
1.4 V (RMS). By adding external series resistors to
provide a suitable attenuation, the external input could be
used as a SCART input. Whenever the external or mono
input is selected, the output of the DAC is muted to
improve the crosstalk performance.
6.3.3D
UAL AUDIO DAC
The TDA9874H comprises a two-channel audio DAC for
feeding signals from the DSP section to the analog
crossbar switch. These DACs have a resolution of 15 bits
and employ four-fold oversampling and noise shaping.
6.3.4A
UDIO OUTPUT BUFFERS
The output buffers provide 0 dB of gain and offer a muting
possibility. The post filter capacitors of the audio DACs are
connected to the buffer outputs.
6.3.5S
TANDBY MODE
The Standby mode (see Section 7.3.2) disables most
functions and reduces power dissipation of the
TDA9874H, but provides no other functionality.
Internal registers may lose their information in Standby
mode. Therefore, the device needs to be initialized on
returning to normal operation. This can be accomplished in
the same way as after a power-on reset.
handbook, full pagewidth
mono (AM)
EXTIL
EXTIR
DACL
DACR
source select
Fig.4 Switch diagram for the audio section.
1998 Apr 2714
matrix
OUTL
OUTR
MGK754
This text is here in white to force landscape pages to be rotated correctly when browsing through the pdf in the Acrobat reader.This text is here in
_white to force landscape pages to be rotated correctly when browsing through the pdf in the Acrobat reader.This text is here inThis text is here in
white to force landscape pages to be rotated correctly when browsing through the pdf in the Acrobat reader. white to force landscape pages to be ...
1998 Apr 2715
mono
handbook, full pagewidth
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
external
NICAM
FM/AM
NICAM
DEMODULATOR
FM/AM
DEMODULATOR
NICAM
DECODER
ADAPTIVE
DE-EMPHASIS
DE-EMPHASIS
FIXED
DE-EMPHASIS
LEVEL
ADJUST
2 CHANNEL
DECODER
LEVEL
ADJUST
Fig.5 Audio signal flow.
DIGITAL
CROSSBAR
SELECT
LEVEL
ADJUST
DAC
ANALOG
CROSSBAR
SWITCH
MATRIX
MATRIX
BUFFER
LEVEL
ADJUST
OUT
I
MGK756
2
S
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
7I2C-BUS CONTROL
7.1Introduction
2
The TDA9874H is controlled only via the I
C-bus. Control
is exercised by writing data to one or more internal
registers. Status information can be read from an array of
registers to let the controlling microprocessor determine
whether any action is required.
The device has an I2C-bus slave transceiver in
accordance with the fast-mode specification with a
maximum speed of 400 kbits/s. Information about the
I2C-bus can be found in brochure
it”
(order number 9398 393 40011). To avoid conflicts in a
“I2C-bus and how to use
real application with other ICs providing similar or
complementing functions, there are four possible slave
addresses available, which can be selected by pins
ADDR1 and ADDR2 (see Table 7).
Table 7 Possible slave addresses
SLAVE ADDRESS
ADDR2ADDR1
A6 A5 A4 A3 A2 A1 A0
001011000
011011001
101011010
111011011
2
The I
C-bus interface remains operational in the Standby
mode of the TDA9874H to allow the device to be
reactivated via the I2C-bus.
7.2Power-up state
At power-up the device is in the following state:
• All outputs muted
• No sound carrier frequency loaded
• General purpose I/O pins ready for input (HIGH)
• Input SIF1 selected with:
– AGC on
– Small hysteresis.
• Demodulators for both sound carriers set to FM with:
– Identification for B/G, D/K, identification mode ‘slow’
– Level adjust set to 0 dB
– De-emphasis 50 µs
– Dematrix set to mono
– Adaptive de-emphasis on.
• OUTL and OUTR set to mono and connected to DAC
• Digital audio interface all outputs off
• Monitor set to carrier 1 DC output.
After power-up a device initialization has to be performed
2
via the I
C-bus to put the TDA9874H into the proper mode
of operation, in accordance with the desired TV standard,
etc. This can be done by writing to all registers with a
single I2C-bus transmission (like a refresh operation) or by
writing selectively only to those registers, the contents of
which need to be changed with regard to the power-up
state.
The device will not respond to a ‘general call’ on the
I2C-bus, i.e. when a slave address of 0000000 is sent by a
master.
1998 Apr 2716
Philips SemiconductorsPreliminary specification
Digital TV sound demodulator/decoderTDA9874H
7.3Slave receiver mode
As a slave receiver, the TDA9874H provides 24 registers for storing commands and data. Each register is accessed via
a so-called subaddress. A subaddress can be thought of as a pointer to an internal memory location.
Detailed descriptions of the slave receiver registers are given in Sections 7.3.1 to 7.3.20.
2
Table 8 I
SSLAVE ADDRESS0ASUBADDRESSADATAA/NA P
Table 9 Explanation of Table 8
SSTART condition
SLAVE ADDRESS
0data direction bit (write to device)
A
SUBADDRESS
DATAdata byte to be written into register
A/NAacknowledge or not acknowledge
PSTOP condition
C-bus; SLAVE ADDRESS/SUBADDRESS/DATA format
BITFUNCTION
7-bit device address
acknowledge
address of register to write to
It is allowed to send more than one data byte per transmission to the TDA9874H. In that case, the subaddress is
automatically incremented after each data byte, resulting in storing the sequence of data bytes at successive register
locations, starting at SUBADDRESS. A transmission can start at any valid subaddress. Each byte that is properly stored,
is acknowledged with A (acknowledge). If an attempt is made to write data to a non-existing subaddress, the device
acknowledges with NA (not acknowledge), therefore telling the I
‘wrap-around’ of subaddresses.
Commands and data will be processed as soon as they have been received completely. Functions requiring more than
one byte will, thus, be executed only after all bytes for that function have been received. If the transmission is terminated
(STOP condition) before all bytes have been received, the incomplete data for that function are ignored.
Table 10 Format for a transmission employing auto-increment of subaddresses
S SLAVE ADDRESS 0 A SUBADDRESS ADATA BYTE A
Data patterns sent to the various subaddresses are not checked for being illegal or not at that address, except for the
level adjust functions.
Detection of a STOP condition without a preceding acknowledge bit is regarded as a bus error. In this case, the last
operation will not be executed.
2
C-bus master to abort the transmission. There is no
DATA A/NA P
n data bytes with auto-increment of
subaddresses
1998 Apr 2717
Loading...
+ 39 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.