PHILIPS SAA2521

INTEGRATED CIRCUITS
DATA SHEET
SAA2521
Masking threshold processor for
MPEG layer 1 audio compression
applications
Preliminary specification
File under Integrated Circuits, IC01
August 1993
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
FEATURES
GENERAL DESCRIPTION
• Stereo or 2-channel mono encoding
The SAA2521 performs the adaptive allocation and
scaling function for calculating the masking thresholds and
sub-band sample accuracy in MPEG layer 1 applications.
The SAA2521 is intended for use in conjunction with the
stereo filter codec SAA2520.
• Status may be read continuously
• Microcontroller interface
• I2S-interfaces
• Allocation algorithm including optional emphasis
correction (for 44.1 kHz)
• Reduced power consumption
• 4 V nominal operating voltage capability.
ORDERING INFORMATION
EXTENDED TYPE
NUMBER
SAA2521GP
PACKAGE
PINS
PIN POSITION
MATERIAL
CODE
44
QFP
plastic
SOT205AG(1)
Note
1. SOT205-1; 1996 August 23.
August 1993
2
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
VDD
handbook, full pagewidth
SWS
FS256
14,24,40
31
39
33
FDAF
NODONE
RESOL0
RESOL1
FDIR
FRESET
FSYNC
SCALE
TEST3
TEST4
LTDATA
LTCNT1
LTCNT0
LTENA
LTCLK
CLK24
RESET
PWRDWN
34
INTERFACE
COMPENSATION DELAY
FDAC
SCL
20
21
22
37
36
SAA2521
35
38
ALLOCATION AND
SCALING
CALCULATION
CONTROL
15
16
5
1
2
3
4
11
7
8
9
10
LT INTERFACE
26
23
30
6,25,44
VSS
Fig.1 Block diagram.
August 1993
32
INTERFACE
3
MLB137
LTDATAC
LTCNT1C
LTCNT0C
LTENC
LTCLKC
Philips Semiconductors
Preliminary specification
FDAF
34
FRESET
36
35 FSYNC
FDIR
37
V DD
40
SAA2521
38 SCALE
n.c.
41
39 FS256
n.c.
n.c.
42
handbook, full pagewidth
43
VSS
44
Masking threshold processor for MPEG
layer 1 audio compression applications
LTCNT1
1
33 FDAC
LTCNT0
2
32 SCL
LTENA
3
31 SWS
LTCLK
4
30 PWRDWN
LTDATA
5
29 TEST10
V SS
6
LTCNT1C
7
27 TEST8
LTCNT0C
8
26 CLK24
LTENC
9
25 V SS
SAA2521
28 TEST9
24 V DD
LTCLKC 10
23 RESET
14
15
16
17
18
19
20
21
22
TEST3
TEST4
TEST5
TEST6
TEST7
NODONE
RESOL0
RESOL1
13
TEST2
V DD
12
TEST1
LTDATAC 11
MLB136
Fig.2 Pin configuration.
handbook, full pagewidth
AUDIO
SOURCE
AUDIO
AMPLIFIER
digital audio interface
ADC/DAC
control
SAA2520
and
SAA2521
system micro interface
MICROCONTROLLER
MPEG interface
power down
reset
MLB138
Fig.3 MPEG codec system data flow diagram.
August 1993
MPEG
source/
receiver
4
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
PINNING
SYMBOL
PIN
DESCRIPTION
TYPE
LTCNT1
1
mode control 1, microcontroller interface input
I
LTCNT0
2
mode control 0, microcontroller interface input
I
LTENA
3
enable microcontroller interface input
I
LTCLK
4
bit clock microcontroller interface input
I
LTDATA
5
data, microcontroller interface (3-state inputs/outputs)
I/O
VSS
6
supply ground (0 V)
LTCNT1C
7
control 1; microcomputer interface
O
LTCNT0C
8
control 0; microcomputer interface
O
LTENC
9
enable microcontroller interface
O
LTCLKC
10
bit clock; microcontroller interface
O
LTDATAC
11
data; microcontroller interface, (3-state inputs/outputs)
I/O
TEST1
12
test output; do not connect
TEST2
13
test output; do not connect
VDD
14
positive supply voltage (+ 5 V)
TEST3
15
test mode input; to be connected to VDD
TEST4
16
test mode input; to be connected to VDD
TEST5
17
test input; to be connected to VSS
TEST6
18
test input; to be connected to VSS
TEST7
19
test input; to be connected to VSS
NODONE
20
no done state selection input
I
RESOL0
21
resolution selection 0 input
I
RESOL1
22
resolution selection 1 input
I
RESET
23
active HIGH reset input
I
VDD
24
positive supply voltage (+ 5 V)
VSS
25
supply ground (0 V)
CLK24
26
24.576 MHz processing clock input
TEST8
27
test input; to be connected to VSS
TEST9
28
test input; to be connected to VSS
TEST10
29
test input; to be connected to VSS
PWRDWN
30
power-down input
SWS
31
word selection input; (Filtered) - I2S-interface
SCL
32
I
I
bit clock input; (Filtered) -
I
I2S-interface
I2S-interface
I
FDAC
33
filtered data (Filtered) -
(3-state inputs/outputs)
I/O
FDAF
34
filtered data (Filtered) - I2S-interface (3-state inputs/outputs)
I/O
I2S-interface,
input
I
FSYNC
35
sub-band synchronization on (Filtered) -
FRESET
36
reset signal input from SAA2520
I
FDIR
37
direction of the I2S-interface; input
I
SCALE
38
scale factor index select (note 1)
I
FS256
39
system clock input; sample frequency × 256
I
VDD
40
positive supply voltage (+ 5 V)
August 1993
5
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SYMBOL
PIN
SAA2521
DESCRIPTION
n.c.
41
not connected
n.c.
42
not connected
n.c.
43
not connected
VSS
44
supply ground (0 V)
TYPE
Note to the Pinning Description
1. The scale input must be set LOW for use with the SAA2521.
FUNCTIONAL DESCRIPTION
Encoding Mode
Coding System
Signal FDIR sets the data flow direction on the
Filtered-I2S-interface. In the encoding mode (FDIR LOW)
the device will accept samples from FDAF. These will be
delayed by a number of sample periods depending upon
the setting of the SCALE input. In the instance of operation
with the SAA2520 (SCALE = logic 0) this delay will be 480
SWS periods. This will ensure alignment of the data with
the computed allocations.
This efficient MPEG audio encoder is used in conjunction
with the SAA2520 filter codec (bit rates of 384, 256, 192
and 128 k bits/s). The encoder utilizes a system producing
sub-band samples from an incoming digital audio signal.
This relies upon the audibility of signals above a given
level and upon high amplitude signals masking those of
lower amplitude. Although each sub-band signal is of
approximately 750 Hz bandwidth, it possesses
considerable overlap with those adjacent to it.
After the delay the samples will be presented on FDAC
(pin 33). The circuit also performs all the calculations
required to build the allocation table which is used in the
codec (SAA2520). When used with the SAA2520 the
calculated scale factor indices are sent via the LT
interface. These operations are performed for every frame
of the sub-band codec.
During the process of encoding, the masking threshold
processor analyses the broadband audio signal at
sampling frequency fs by splitting it into 32 sub-band
signals at a sampling frequency (fs/32).
The coded signal consists of frames conveying the
information corresponding to the sub-band samples.
These also include a synchronization pattern identifying
the start of each new frame. The allocation information for
the 32 sub-bands is transferred as 4-bit values. If the
amplitude of a sub-band signal is below the masking
threshold it will be omitted from the coded signal.
In order to synchronize with the codec and utilize the
correct tables for the calculations the SAA2521 frequently
requests the status of the codec. It monitors the bit-rate,
sample frequency, operation mode and the emphasis
information and uses the 'ready to receive' bit of the codec
to determine the moment of the transfer of allocation
information.
The duration of a MPEG frame depends upon sampling
frequency and is adjusted to 384 divided by fs.
Decoding Mode
In the decoding mode (FDIR HIGH) the SAA2521 will take
samples from FDAC which will be presented on the FDAF
after a delay of 160 SWS periods. The LT interface
between microcontroller and codec (SAA2520) will only be
affected by the 'ready to receive' bit from the codec
(SAA2520).
Adaptive Allocation and Scaling
The coding system calculates the masking power of the
sub-band signals and adds the masking threshold.
Sub-band signals with power below this threshold denote
information to be discarded. Non-masked signals are
coded using floating point notation in which a mantissa
corresponds in length to the difference between peak
power and masking threshold. The process is repeated for
every MPEG frame and is known as the Adaptive
Allocation of the available capacity.
August 1993
6
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
Mode Control
Operation is controlled by the FRESET and FDIR signals.
FRESET causes a general reset. The FDIR signal is
sampled at the falling edge of the FRESET signal to
determine the operation mode:
Microcontroller Interface Operation
Information on the interface between microcontroller and
codec (SAA2520) will flow in a regular sequence
synchronized with the codec (SAA2520):
- with every FSYNC the SAA2521 will read the status of
the codec (SAA2520)
- Following the calculation of the allocation and scale
factors the SAA2521 will send the first allocation
information unit (16-bits). It will then continuously read
the codec (SAA2520) status to ascertain when it is
able to receive further allocation information units.
When the transfer of these units is complete the
SAA2521 will send settings and (for SCALE = logic 0)
scale factor indices.
FDIR = logic 1
decoding mode, SAA2521 in
feed-through mode
FDIR = 0
encoding mode, SAA2521 in
calculation mode
Fig.4 shows the timing diagram for FRESET and FDIR.
Resolution Selection
The (SAA2521) is designed for operation with input
devices (ADCs) which may possess a different sample
resolution capability, i.e. audio sample inputs into the
sub-band filters. Pins RESOL0 and RESOL1 (respectively
pins 21 and 22) may be utilized to adjust the allocation
information calculation to the resolution of the samples.
- The extended settings will be sent to the codec as
soon as possible after reception from the
microcontroller.
The microcontroller communicates with the SAA2521 in
a similar fashion:
- status can be read continuously. The SAA2521 will
output a copy of the codec (SAA2520) status on the
LTDATA line except for the 'ready to receive' bits
which are generated by the SAA2521. These indicate
whether the SAA2521 is ready to receive the next
settings or extended settings.
With the instance of pin 20 (NODONE) being HIGH, all
available bits in the bit-pool will be allocated. If NODONE
is LOW, no bits will be allocated to the sub-bands with
energy levels below the theoretical threshold for the
selected resolution.
- settings can be sent following every occasion that the
'ready to receive' bit 'S' changes to logic 1.
- extended settings can be sent following each
occasion that the 'ready to receive' bit 'E' changes to
logic 1.
t rH
FRESET
t suD
FDIR
MBC123 - 1
TrH > 5TCLK24 = 210 ns (for CLK24 = 24.576 MHz) min. time FRESET HIGH
TsD < 0 ns
min. set-up time FDIR to FRESET = LOW
Fig.4 Timing: FRESET and FDIR.
August 1993
7
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
Power-down Mode Switching
SAA2521
Table 1
When the potential on the RESET pin (pin 23) is held HIGH
for at least 5TCLK24 clock periods, the device will be reset
after which it will operate in its decoding mode.
Resolution selection.
RESOL1
RESOL0
RESOLUTION
0
0
16-bits
0
1
18-bits
1
0
14-bits
1
1
15-bits
The power-down mode is activated when the PWRDWN
pin (pin 30) is held HIGH. The 3-state buffers will be set to
a high impedance while the normal outputs will retain the
state attained prior to this mode being entered. This mode
can only be used if other associated circuits react
accordingly. The power-down mode is de-activated by a
reset action.
Fig.5 shows the operation for the power-down mode
switching.
PWRDWN
sleep mode active
RESET
t rH
MEA659 - 1
TrH > 5TCLK24 = 210 ns (for CLK24 = 24.576 MHz) minimum time RESET HIGH
Fig.5 Power-down mode switching.
August 1993
8
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
channel
SAA2521
right
left 32 bits
SWS
SCL
1
7 bits
FDA
bit :
2
3
2
2
2
1
2
0
2
1
0
msb
2
3
msb
lsb
2
2
2
1
2
0
MBC149 - 1
Fig.6 Format for transferring filtered data.
channel
L
R
L
R
L
R
L
R
L
R
L
R
L
R
SWS
FSYNC
sub-band
31
0
1
31
0
1
MBC126 - 2
Fig.7 FSYNC related to SWS 0 data transfer period.
August 1993
9
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
Table 2
The (Filtered) - I2S-interface.
SWS
input
word selection
Fs
SCL
input
bit clock
64 Fs
FDAF
bi-directional
filtered data to/from the filter section of SAA2520
FDAC
bi-directional
filtered data to/from the codec section of SAA2520
FSYNC
input
filter synchronization
Table 3
The (Filtered) - I2S-interface.
FRESET
input
reset
FDIR
input
Filtered - I2S-interface direction of data flow
(Filtered) - I2S-interfaces
Interfaces with the sub-band filter and codec (SAA2520)
consist of the following signals.
Fig.6 shows the format for transferring filtered data.
Fs 256 must be provided as system clock. This frequency
is used by the interfaces with the SAA2520.
The frequency of the SWS signal (pin 31) is equal to the
sample frequency Fs. Bit clock SCL (pin 32) is 64 times the
sample frequency; thus each SWS period contains 64 data
bits, 48 of which are actually used in data transfer. The half
period during which SWS is logic 0 is used to transfer
Left-channel information while that during which it is 1
permits transfer of Right-channel data.
The 24-bit samples are transferred with the most
significant bit first. This bit is transferred during the bit clock
period, one bit time after the change in SWS.
FSYNC signal is provided for the purposes of
synchronization and indicates the portion of the SWS
period during which the samples of sub-band 0 are
transferred.
Fig.7 shows the relationship between FSYNC and the
SWS 0 data transfer period.
August 1993
10
Fs/32
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
4T
t cH
t cL
SCL
t d4
t d3
output
t su1
t h1
input
MBC127 - 2
OUTPUT applies to FDAF and FDAC in the output mode.
INPUT applies to FDAF and FDAC in the input mode, SWS and FSYNC.
T = one Fs256 cycle time
minimum HIGH time SCL
tcH ≥ T + 35 ns
TcL ≥ T + 35 ns minimum LOW time SCL
td3 ≥ 2T − 10 ns hold time output after SCL HIGH
td4 ≤ 3T + 60 ns delay time output after SCL HIGH
set-up time input before SCL HIGH
ts1 ≥ 20 ns
th1 ≥ T + 35 ns hold time input after SCL HIGH
FDIR
t W1
FDA
t W2
HIGH Z
HIGH Z
MEA692 - 1
tw1 ≥ 3T minimum time high impedance to FDA enabled
tw2 ≥ 2T + 35 ns maximum time FDA enabled to high impedance
Fig.8 (Filtered) - I2S-interface timing.
August 1993
11
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
Table 4
SAA2521 interface with microcontroller.
LTCLK
input
LTDATA
LTCNT0
SAA2521
Table 5
SAA2521 interface with SAA2520.
bitclock
LTCLKC
output
bi-directional
data
LTDATAC
bi-directional
data
input
control line 0
LTCNT0C
output
control line 0
LTCNT1
input
control line 1
LTCNT1C
output
control line 1
LTENA
input
enable
LTENC
output
enable
Table 6
bit clock
SAA2521 interface control lines functions.
LTCNT1(C)
LTCNT0(C)
MODE
FROM
TO
TRANSFER OF
0
0
extended settings
microcontroller
SAA2520
8-bits
0
1
allocation (see note)
SAA2521
SAA2520
16/48 × 16-bits
1
0
settings
microcontroller
SAA2520
16-bits
1
1
status
codec
microcontroller
8 or 16-bits
A transfer of information begins when the master sets the
control lines for the required action. It then sets the
LTENA/C line to logic 1. Once this signal is established the
slave determines the kind of action required and prepares
for the transfer of data.
Microcontroller Interface
Two microcontroller interfaces are provided; one for
connection to the microcontroller interface of the
SAA2520, the other to connect to the system controller.
Information is conveyed via the SAA2521 which executes
monitoring and extracts signals (e.g. settings and
synchronization) essential to its operation. Additionally it
also sends allocation information to the SAA2520.
However, the SAA2521 does not monitor the external
settings bits from the microcontroller (see Extended
Settings).
When the master supplies the LTCLK/C signal, data is
transferred either to or from the slave in units of 8-bits; the
least significant bit is always transferred first. A transfer of
16-bits is made in two, 8-bit units with the most significant
8-bit unit first. In between the two 8-bit units the LTENA/C
signal remains logic 1.
Fig.9 shows an example of information transfer via
SAA2521 interfaces.
The SAA2521 is a slave on the interface with the
microcontroller which is active only when the enable signal
LTENA (pin 3) is logic 1. This permits connection of this
interface to other devices. Only the enable signal is not
common to all devices.
Note to Table 6
This mode only on the interface between SAA2521 and
SAA2520.
SAA2521 is master on the interface with the SAA2520 and
provides all signals with the exception of the data in the
instance of status transfer from SAA2520 to SAA2521.
If SCALE = logic 1 then 16 × 16-bits
If SCALE = logic 0 then 48 × 16-bits
Information conveyed via these interfaces is transferred in
8 or 16-bit serial units with the type of information
designated by the control lines (LTCNT1(C) and
LTCNT0(C)).
August 1993
12
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
LTENA/C
LTCNT0(C)/1(C)
LTCLK(C)
LTDATA(C)
0
1
2
3
4
5
6
7
MBC128
Fig.9 Example of information transfer via SAA2521 interfaces.
LTENA/C
LTCNT0(C)/1(C)
LTCLK(C)
LTDATA(C)
E0
E1
E2
E3
E4
E5
E6
Refer to the SAA2520 description for the meaning of these bits as they pass SAA2521 unchanged.
Fig.10 Extended settings (LTCNT1 and LTCNT0)
August 1993
13
E7
MBC129
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
The information will consist of 16 transfers each of 16-bits.
To synchronize the SAA2521 operation with that of the
SAA2520, following the first 16-bit transfer of allocation
data the SAA2521 checks the SAA2520 status to ensure it
is ready to receive the remainder of the allocation
information. Transfer of allocation data is completed by
sending settings. Between 16-bit transfers the LTENC line
returns to 0 as shown in Fig.11.
Extended settings (LTCNT1(C) = logic 0, LTCNT0(C) =
logic 0)
Eight information bits, generated by the microcontroller,
are transferred in this mode. The SAA2521 will transfer
these bits to the SAA2520 as soon as possible but does
not monitor this information.
Fig.10 shows the relationship of the extended settings.
Fig.12 shows the order in which the bits occur on the
interface during allocation information transfer.
Allocation and SCALING information (LTCNT1C =
logic 0, LTCNT0C = logic 1)
The 4-bit sub-band allocation unit contains the number of
bits allocated to the sub-band MINUS 1. A value of 0000
indicates no bits allocated to that sub-band.
In the encoding mode (FDIR = logic 0) the SAA2521 will
transfer allocation information to the SAA2520. This will
occur once for every SAA2520 frame.
LTENC
16 bits
16 bits
LTCLKC
MBC130
Fig.11 LTENC behaviour for 16-bit transfers.
Table 7
Allocation and SCALING information.
MSB
BITS
LSB
CHANNEL
SUB-BAND
A15
−
A14
−
A13
−
A12
L
2*
COUNT
A11
−
A10
−
A9
−
A8
R
2*
COUNT
A7
−
A6
−
A5
−
A4
L
(2 *
COUNT) + 1
A3
−
A2
−
A1
−
A0
R
(2 *
COUNT) + 1
Table 8
Allocation and SCALING information.
MSB
BITS
LSB
SL15
−
SL14
−
SL13
−
SL12 - SL11 - SL10 - SL9
−
SL8
SL7
−
SL6
SL5
−
SL4 - SL3 - SL2 - SL1
−
SL0
August 1993
14
CHANNEL
CONTENTS
---
00
L
SCALE FACTOR (COUNT)
---
00
R
SCALE FACTOR(COUNT)
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
With stereo encoding, Left and Right channels are
designated L and R. This changes to channels I or II for
2-channel mono mode. If SCALE = logic 0 the transfer of
allocation information will be followed by the transfer of
scale factors. Each 16-bit transfer contains two scale
factor indices.
Algorithm showing the process of information transfer:
COUNT: = logic 0
SEND ALLOCATION (COUNT)
REPEAT
READ STATUS
UNTIL
READY-TO-RECEIVE
FOR COUNT: = 1 to 15
DO
SEND ALLOCATION (COUNT)
SEND SETTINGS
IF SCALE = logic 0
THEN
FOR COUNT; = logic 0 TO 31
DO
SEND SCALE FACTORS (COUNT)
LTENA/C
LTCNT0(C)/1(C)
LTCLK(C)
LTDATA(C)
bit :
A or SL :
8
9
1
0
1
1
1
2
1
3
1
4
1
5
0
1
2
3
4
5
6
7
Fig.12 Order of interface bits during allocation information transfer.
August 1993
15
MEA691
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
SETTINGS (LTCNT1(C) = logic 1, LTCNT0(C) = logic 0)
Without using the information, the SAA2521 transfers
microcontroller settings to the SAA2520.
Prior to sending settings, the microcontroller would utilize
the SAA2521 status readings to ensure its readiness to
accept and convey the data.
Following reception of the settings the SAA2521 will cause
the ready-to-receive bit to be logic 0 until the settings have
been sent to the SAA2520. The microcontroller can only
send this data when this bit is logic 1.
Fig.13 shows the order of the bits on the interface.
LTENA/C
LTCNT0(C)/1(C)
LTCLK(C)
LTDATA(C)
S:
bit :
8
9
1
0
1
1
1
2
1
3
1
4
1
5
0
1
2
3
4
5
6
7
MBC132
Fig.13 The order of bits on the interface.
Table 9
Microprocessor settings applied to the SAA2520 via the SAA2521.
MSB
BITS
S15
−
S14
S11
−
S10
S9
−
LSB
S13
−
S12
NAME
FUNCTION
VALID IN
bitrate index
bitrate indication
encode
sample frequency
44.1, 48 or 32 kHz indic.
encode
DECODE
1 - decode; 0 - encode
enc/dec
S8
ext 256fs
1 - ext; 0 - int
enc/dec
S7
2-ch mono
1 - 2 ch mono; 0 - stereo
encode
S6
MUTE
1 - mute; 0 - no mute
enc/dec
S5
not used
−
enc/dec
S4
CH I
1 - CH I; 0 - CH II
decode
S3
−
S2
Tr0 - Tr1
transparent bits
encode
S1
−
S0
EMPHASIS
emphasis indication
encode
August 1993
16
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
The SAA2521 can only be used to encode stereo (mode
00) signals and 2-channel mono (mode 10) signals.
Status (LTCNT1(C) = logic 1, LTCNT0(C) = logic 1)
The SAA2520 and SAA2521 operation may be checked by
reading these bits. All, except the ready-to-receive bits,
are generated by the SAA2520.
During the decoding mode this bit indicates if the operation
of the SAA2520 is in synchronization with the MPEG
coded signal. Should this not be the case the SAA2520
cannot perform the decoding.
The bit rate index indicates the bit rate of the sub-band
signal in units of 32 kbits/s. The SAA2521 is designed for
bit rates of 384, 256, 192 and 128 kbits/s only.
CLKOK indicates whether or not the Fs256 clock
corresponds with the specified sample frequency.
With EMPHASIS activated (S1 = T1 = 0 and S0 = T0 = 1)
only bit rates 384 and 256 kbits/s can be used.
EMPHASIS indication may be used to apply correct
de-emphasis. During the encoding 50 / 15 µs mode the
SAA2521 will correct the calculated allocation if emphasis
is applied for a 44.1 kHz sampling frequency.
A ready-to-receive S or E indicates whether or not the
SAA2521 can receive new settings or extended settings
respectively from the microcontroller and should be
checked prior to sending new information.
Table 10 Order of SAA2520 bits as they appear on the interface (see also Fig.14).
MSB
BITS
T15
-
T14
T11
-
T10
−
LSB
T13
−
T12
NAME
bitrate index
FUNCTION
bitrate indication
VALID IN
enc/dec
sample frequency 44.1, 48 or 32 kHz indic.
enc/dec
T9
ready-to-rec S
1 - ready; 0 - not ready
enc/dec
T8
ready-to-rec E
1 - ready; 0 - not ready
enc/dec
MODE
sub-band signal mode ID
enc/dec
T5
SYNC
synchronization indic.
dec
T4
CLKOK
1 - OK; 0 - not OK
enc/dec
T7
−
T6
T3
−
T2
Tr0 - Tr1
transparent bits
enc/dec
T1
−
T0
EMPHASIS
emphasis indication
enc/dec
August 1993
17
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
LTENA/C
LTCNT0(C)/1(C)
LTCLK(C)
LTDATA(C)
T:
bit :
8
9
1
0
1
1
1
2
1
3
1
4
1
5
0
1
2
3
4
5
6
7
MBC133
Fig.14 Order of appearance of bits on the interface.
Sample frequency indication.
MSB
Frequency Range Limitation
In encode mode the frequency range will be limited at
lower rates. This is implemented by making the samples of
higher frequency sub-bands equal to logic 0 before the
allocation calculation. This automatically ensures that
these sub-bands do not get any bits allocated.
LSB
00
44.1 kHz
default value
01
48 kHz
10
32 kHz
11
--
do not use
MODE
OUTPUT
The following table shows the sub-bands affected and the
resulting frequency range.
MODE identification.
MSB LSB
00
stereo
L and R
01
joint stereo
L and R
10
2 - channel
mono
I or II as selected
11
1 - channel
mono
mono, no selection
August 1993
The transfer of either 8-bits or 16-bits is permitted for the
transfer of status information. When only 8-bits are
transferred, these will always form the first byte and may
be used in checking the ready-to-receive bit.
18
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
Table 11 Frequency examples.
BIT RATE
FS
SUB-BANDS SET TO 0
@ FREQUENCY
256 kbit/s
48 kHz
29, 30, 31
> 21750 Hz
192 kbit/s
48 kHz
20, 21, ... , 30, 31
> 15000
44.1 kHz
22, 23, ... , 30, 31
> 15159
48 kHz
12, 13, ... , 30, 31
> 9000
44.1 kHz
13, 14, ... , 30, 31
> 8957
32 kHz
20, 21, ... , 30, 31
>10000
128 kbit/s
t Le
LTENA
t su1
t h2
t h1
LTCNT0/1
t su4
t su2
t cL
t cH
LTCLK
t su3
t h3
LTDATA
0
bit :
tle
ts1
th1
ts2
th2
> 210 ns minimum LOW time LTENA prior to transfer
> 50 ns set-up time LTCNT0, 1 before LTENA HIGH
> 210 ns hold time LTCNT0, 1 after LTENA HIGH
> 210 ns set-up time LTENA before LTCLK LOW
> 210 ns hold time LTENA after LTCLK HIGH
tlc
thc
ts3
th3
ts4
> 210 ns minimum LOW time LTCLK
> 210 ns minimum HIGH time LTCLK
> 210 ns set-up time LTDATA before LTCLK HIGH
> 50 ns hold time LTDATA after LTCLK HIGH
> 210 ns set-up time LTCLK before LTENA HIGH
1
Fig.15 Microcontroller to SAA2521 timing.
August 1993
19
MEA658 - 2
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
LTENA
SAA2521
LTENA must remain HIGH
LTCNT0/1
LTCLK
MBC135 - 1
t W1
tw1 > 550 ns
minimum time between two 8-bit transfers
Fig.16 16-bit transfers.
t Le
LTENA
t su1
t h2
t h1
LTCNT0/1
t su4
t su2
t cL
t cH
LTCLK
t d1
t d2
t h5
t h6
LTDATA
0
bit :
tle
ts1
th1
ts2
> 210 ns minimum LOW time LTENA prior to transfer
> 50 ns set-up time LTCNT0, 1 before LTENA HIGH
> 210 ns hold time LTCNT0, 1 after LTENA HIGH
> 210 ns set-up time LTENA before LTCLK LOW
th2
tlc
> 210 ns hold time LTENA after LTCLK HIGH
> 210 ns minimum LOW time LTCLK
thc
td1
td2
th5
tS4
th6
>
<
210 ns minimum HIGH time LTCLK
385 ns maximum delay LTDATA after LTENA HIGH
<
>
>
>
385 ns maximum delay LTDATA after LTCLK HIGH
145 ns hold time LTDATA after LTCLK HIGH
210 ns set-up time LTCLK before LTENA HIGH
0 ns hold time LTDATA after LTENA LOW
1
Fig.17 SAA2521 to Microcontroller timing.
August 1993
20
MEA657 - 2
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
LTENA
SAA2521
LTENA must remain HIGH
LTCNT0/1
LTCLK
MBC137
t W2
tw2 > 550 ns
minimum time between two 8-bit transfers
Fig.18 16-bit transfers.
t Le
LTENC
t h2
t su1
LTCNT0(C)/1(C)
t su4
t su2
t cL
t cH
LTCLKC
t su3
t h3
LTDATAC
0
bit :
tle
ts1
ts2
th2
tlc
thc
ts3
th3
ts4
1
> 400 ns minimum LOW time LTENA prior to transfer
> 400 ns set-up time LTCNT0, 1C before LTENC HIGH
> 200 ns set-up time LTENC before LTCLKC LOW
> 400 ns hold time LTENC after LTCLK HIGH
> 210 ns minimum LOW time LTCLKC
>
>
>
>
210 ns
210 ns
160 ns
900 ns
minimum HIGH time LTCLKC
set-up time LTDATAC before LTCLKC HIGH
hold time LTDATAC after LTCLKC HIGH
set-up time LTCLKC before LTENC HIGH
Fig.19 SAA2521 to SAA2520 timing.
August 1993
21
MBC138 - 2
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
LTENC
SAA2521
LTENC must remain HIGH
LTCNT0(C)/1(C)
LTCLKC
MBC139
t W2
tw2 > 600 ns
minimum time between two 8-bit transfers
Fig.20 16-bit transfers.
t Le
LTENC
t h2
t su1
LTCNT0(C)/1(C)
t su4
t su2
t cL
t cH
LTCLKC
t d1
t d2
t h5
t h6
LTDATAC
0
bit :
tle
ts1
ts2
th2
tlc
thc
td1
td2
ts4
th5
th6
1
> 400 ns minimum LOW time LTENC prior to transfer
> 400 ns set-up time LTCNT0, 1C before LTENC HIGH
> 200 ns set-up time LTENC before LTCLKC LOW
> 400 ns hold time LTENC after LTCLKC HIGH
> 210 ns minimum LOW time LTCLKC
> 210 ns minimum HIGH time LTCLKC
< 300 ns maximum delay LTDATAC after LTENC HIGH
<
>
>
>
300 ns maximum delay LTDATAC after LTCLKC HIGH
900 ns set-up time LTCLKC before LTENC HIGH
160 ns hold time after LTCLKCC HIGH
0 ns hold time LTDATAC after LTENC LOW
Fig.21 SAA2520 to SAA2521 timing.
August 1993
22
MBC140 - 2
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
LTENC
SAA2521
LTENC must remain HIGH
LTCNT0(C)/1(C)
LTCLKC
MBC141
t W2
tw1 > 600 ns
minimum time between two 8-bit transfers
Fig.22 16-bit transfers.
LIMITING VALUES
In accordance with the Absolute Maximum Rating System (IEC 134).
SYMBOL
PARAMETER
MIN.
MAX.
UNIT
VDD
supply voltage
−0.5
6.5
V
VI
input voltage (note 1)
−0.5
VDD + 0.5
V
IDD
supply current
−
100
mA
II
input current
−
± 10
mA
Io
output current
−
± 40
mA
Ptot
total power dissipation
−
550
mW
Tstg
storage temperature
−55
+ 150
°C
Tamb
operating ambient temperature
−40
+ 85
°C
Ves1
electrostatic handling (note 2)
−1500
1500
V
Ves2
electrostatic handling (note 3)
−70
70
V
Notes
1. Input voltage should not exceed 6.5 V unless otherwise specified.
2. Equivalent to discharging a 100 pF capacitor through a 1.5 kΩ series resistor.
3. Equivalent to discharging a 200 pF capacitor through a 0 Ω series resistor.
August 1993
23
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
DC CHARACTERISTICS
VDD = 3.8 to 5.5 V; Tamb = −40 to 85 °C; unless otherwise specified.
SYMBOL
PARAMETER
CONDITIONS
MIN.
TYP.
MAX.
UNIT
Supply
VDD
supply voltage range
3.8
5
5.5
V
IDD
operating current
VDD = 3.8 V
−
15
30
mA
IDD
operating current
VDD = 5 V
−
25
50
mA
IPWRDWN
stand-by current
in power-down
mode
−
100
−
µA
0
−
0.3 VDD
V
Inputs
VIL
LOW level input voltage
VIH
HIGH level input voltage
0.7 VDD
−
VDD
V
II
input current
−
−
10
µA
Outputs
VOL
LOW level output voltage
note 1
−
−
0.4
V
VOH
HIGH level output voltage
note 1
VDD − 0.5
−
−
V
Vi = 0 to 5.5 V
−
−
10
µA
3-state outputs
Ioz
OFF state current
Note
1. Maximum load current for LTDATA, LTCNT1C, LTCNT0C, LTENC, LTCLKC, TEST1, TEST2, FDAC, FDAF = 2 mA;
for LTDATAC = 3 mA.
August 1993
24
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
AC CHARACTERISTICS
VDD = 3.8 to 5.5 V; Tamb = −40 to 85 °C.
SYMBOL
PARAMETER
CONDITIONS
MIN.
TYP.
MAX.
UNIT
CLOCK CLK24
fs
frequency
23
24.576
26
MHz
−
−
13
MHz
CLOCK Fs256
fs
frequency
fs = 48 kHz
Inputs FSYNC, SWS, LTCNT1, LTCNT0, LTENA, LTCLK, LTDATA, LTDATAC, FDAF, FDAC, SCL, SWS
CI
input capacitance
−
−
10
pF
INPUT SET-UP TIME
tSU
set-up time of inputs related to
CLK24 rising edge
note 1
15
−
−
ns
tSU
set-up time of inputs related to
256fs rising edge
note 2
15
−
−
ns
tHD
hold time of inputs related to CLK24 note 1
rising edge
20
−
−
ns
tHD
hold time of inputs related to 256fs
rising edge
10
−
−
ns
INPUT HOLD TIME
note 2
Outputs LTDATA, LTDATAC, LTCNT1C, LTCNT0C, LTENC, LTCLKC, FDAF, FDAC
Co
output capacitance
td
output delay time related to CLK24
rising edge
td
output delay time related to 256fs
rising edge
−
−
10
pF
CL = 25 pF;
note 3
−
−
45
ns
CL = 25 pF;
note 4
−
−
30
ns
3-state outputs
tPHZ
disable time HIGH-to-Z
CL = 25 pF
−
−
65
ns
tPLZ
disable time LOW-to-Z
CL = 25 pF
−
−
65
ns
tPZH
enable time Z-to-HIGH
CL = 25 pF
−
−
65
ns
tPZL
enable time Z-to-LOW
CL = 25 pF
−
−
65
ns
Notes
1. Inputs FSYNC, SWS, LTCNT1, LTCNT0, LTENA, LTCLK, LTDATA, LTDATAC
2. Inputs FDAF, FDAC, SCL, SWS
3. Outputs LTDATA, LTDATAC, LTCNT1C, LTCNT0C, LTENC, LTCLK
4. Outputs FDAF, FDAC
August 1993
25
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
PACKAGE OUTLINE
QFP44: plastic quad flat package; 44 leads (lead length 2.35 mm); body 14 x 14 x 2.2 mm
SOT205-1
c
y
X
33
A
23
34
22
ZE
e
E HE
A
A2
(A 3)
A1
wM
θ
bp
Lp
pin 1 index
44
L
12
detail X
1
11
ZD
e
v M A
wM
bp
D
B
HD
v M B
0
5
10 mm
scale
DIMENSIONS (mm are the original dimensions)
UNIT
A
max.
A1
A2
A3
bp
c
D (1)
E (1)
e
HD
HE
L
Lp
v
w
y
mm
2.60
0.25
0.05
2.3
2.1
0.25
0.50
0.35
0.25
0.14
14.1
13.9
14.1
13.9
1
19.2
18.2
19.2
18.2
2.35
2.0
1.2
0.3
0.15
0.1
Z D (1) Z E (1)
2.4
1.8
2.4
1.8
θ
o
7
0o
Note
1. Plastic or metal protrusions of 0.25 mm maximum per side are not included.
REFERENCES
OUTLINE
VERSION
IEC
SOT205-1
133E01A
August 1993
JEDEC
EIAJ
EUROPEAN
PROJECTION
ISSUE DATE
95-02-04
97-08-01
26
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
SOLDERING
Wave soldering
Introduction
Wave soldering is not recommended for QFP packages.
This is because of the likelihood of solder bridging due to
closely-spaced leads and the possibility of incomplete
solder penetration in multi-lead devices.
There is no soldering method that is ideal for all IC
packages. Wave soldering is often preferred when
through-hole and surface mounted components are mixed
on one printed-circuit board. However, wave soldering is
not always suitable for surface mounted ICs, or for
printed-circuits with high population densities. In these
situations reflow soldering is often used.
If wave soldering cannot be avoided, the following
conditions must be observed:
• A double-wave (a turbulent wave with high upward
pressure followed by a smooth laminar wave)
soldering technique should be used.
This text gives a very brief insight to a complex technology.
A more in-depth account of soldering ICs can be found in
our “IC Package Databook” (order code 9398 652 90011).
• The footprint must be at an angle of 45° to the board
direction and must incorporate solder thieves
downstream and at the side corners.
Reflow soldering
Even with these conditions, do not consider wave
soldering the following packages: QFP52 (SOT379-1),
QFP100 (SOT317-1), QFP100 (SOT317-2),
QFP100 (SOT382-1) or QFP160 (SOT322-1).
Reflow soldering techniques are suitable for all QFP
packages.
The choice of heating method may be influenced by larger
plastic QFP packages (44 leads, or more). If infrared or
vapour phase heating is used and the large packages are
not absolutely dry (less than 0.1% moisture content by
weight), vaporization of the small amount of moisture in
them can cause cracking of the plastic body. For more
information, refer to the Drypack chapter in our “Quality
Reference Handbook” (order code 9397 750 00192).
During placement and before soldering, the package must
be fixed with a droplet of adhesive. The adhesive can be
applied by screen printing, pin transfer or syringe
dispensing. The package can be soldered after the
adhesive is cured.
Maximum permissible solder temperature is 260 °C, and
maximum duration of package immersion in solder is
10 seconds, if cooled to less than 150 °C within
6 seconds. Typical dwell time is 4 seconds at 250 °C.
Reflow soldering requires solder paste (a suspension of
fine solder particles, flux and binding agent) to be applied
to the printed-circuit board by screen printing, stencilling or
pressure-syringe dispensing before package placement.
A mildly-activated flux will eliminate the need for removal
of corrosive residues in most applications.
Several techniques exist for reflowing; for example,
thermal conduction by heated belt. Dwell times vary
between 50 and 300 seconds depending on heating
method. Typical reflow temperatures range from
215 to 250 °C.
Repairing soldered joints
Fix the component by first soldering two diagonallyopposite end leads. Use only a low voltage soldering iron
(less than 24 V) applied to the flat part of the lead. Contact
time must be limited to 10 seconds at up to 300 °C. When
using a dedicated tool, all other leads can be soldered in
one operation within 2 to 5 seconds between
270 and 320 °C.
Preheating is necessary to dry the paste and evaporate
the binding agent. Preheating duration: 45 minutes at
45 °C.
August 1993
27
Philips Semiconductors
Preliminary specification
Masking threshold processor for MPEG
layer 1 audio compression applications
SAA2521
DEFINITIONS
Data sheet status
Objective specification
This data sheet contains target or goal specifications for product development.
Preliminary specification
This data sheet contains preliminary data; supplementary data may be published later.
Product specification
This data sheet contains final product specifications.
Limiting values
Limiting values given are in accordance with the Absolute Maximum Rating System (IEC 134). Stress above one or
more of the limiting values may cause permanent damage to the device. These are stress ratings only and operation
of the device at these or at any other conditions above those given in the Characteristics sections of the specification
is not implied. Exposure to limiting values for extended periods may affect device reliability.
Application information
Where application information is given, it is advisory and does not form part of the specification.
LIFE SUPPORT APPLICATIONS
These products are not designed for use in life support appliances, devices, or systems where malfunction of these
products can reasonably be expected to result in personal injury. Philips customers using or selling these products for
use in such applications do so at their own risk and agree to fully indemnify Philips for any damages resulting from such
improper use or sale.
August 1993
28