S3045B [AMCC]

Mux/Demux, 1-Func, Bipolar, PQFP208, 28 X 28 MM, 3.40 MM HEIGHT, PLASTIC, QFP-208;
S3045B
型号: S3045B
厂家: APPLIED MICRO CIRCUITS CORPORATION    APPLIED MICRO CIRCUITS CORPORATION
描述:

Mux/Demux, 1-Func, Bipolar, PQFP208, 28 X 28 MM, 3.40 MM HEIGHT, PLASTIC, QFP-208

ATM 异步传输模式 电信 电信集成电路
文件: 总30页 (文件大小:206K)
中文:  中文翻译
下载:  下载PDF数据表文档文件
®
DEVICE  
SPECIFICATION  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
APPLICATIONS  
FEATURES  
• SONET/SDH-based transmission systems  
• SONET/SDH modules  
• Complies with Bellcore and ITU-T  
specifications  
• Supports STS-12/STM-4 to STS-48/STM-16  
Mux/Demux functions  
• 8-bit LVDS data path for STS-48/STM-16 data  
• 8-bit LVTTL data path with parity for each  
STS-12/STM-4 data stream  
• SONET/SDH test equipment  
• ATM over SONET/SDH  
• Section repeaters  
• Add drop multiplexers  
• Broad-band cross-connects  
• Fiber optic terminators  
• Optionally calculates even or odd parity over  
parallel data bus or data and frame pulse  
• Compatible with AMCC S3041/S3042 Mux/  
Demux chipset  
• Compatible with PMC PM5355 User Network  
Interface device and PMC PM5312 STTX  
• Optionally calculates and inserts Byte  
Interleaved Parity (B1)  
• Fiber optic test equipment  
• ATM Switch Backbones  
GENERAL DESCRIPTION  
The S3045 SONET/SDH byte interleave chip is a  
fully integrated STS-12/STM-4 to STS-48/STM-16  
Mux/Demux device. The S3045 performs all neces-  
sary byte interleave and byte de-interleave functions  
for multiplexing and de-multiplexing of four STS-12/  
STM-4 data streams into/from a single STS-48/STM-  
16 data stream. The S3045 functions in conformance  
with SONET/SDH transmission standards and is suit-  
able for SONET-based ATM applications. Figure 1  
shows a typical network application. Byte Interleave  
parity (B1) is calculated and inserted for the transmit  
path and calculated, compared and inserted for the  
receive path. Optional frame synchronous scram-  
bling and descrambling are performed, and an  
STS-12/STM-4 framing signal is provided to the  
STS-12/STM-4 interface processors to allow syn-  
chronization of the receive STS-12/STM-4 data  
streams.  
• Optionally compares B1 Byte Interleave Parity  
on the receive side. Generates and inserts B1  
Error indications (B1ERR)  
• Optionally calculates and inserts M1 Bytes, and  
recalculates and inserts the B2 parity bytes  
due to the M1 insertions  
• Optionally inserts section-trace bytes (J0/Z0) in  
the transmit path  
• Diagnostic Loopback Mode  
• Out of Frame (OOF) monitor and alarm indication  
• Loss of Signal (LOS) monitor and alarm indication  
• Squelch Mode: Provides downstream clock  
during Clock Recovery Failure  
• Receive J0 Frame Pulse (J0FP) indicator  
• Performs optional Frame synchronous  
scrambling and descrambling  
• Provides synchronization signal to STS-12/  
STM-4 Network Interface Processors  
• Single 3.3V supply  
• 5 Volt tolerant input  
• 208-pin PQFP/TEP package  
Figure 1. System Block Diagram  
8
8
8
8
8
8
8
8
S3040  
CDR  
S3041  
Tx  
S3042  
Rx  
OTX  
ORX  
8
8
8
8
8
8
8
8
8
S3042  
Rx  
S3040  
CDR  
8
ORX  
S3041  
Tx  
8
8
OTX  
1
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
7. Frame synchronous scrambling.  
8. B1 calculation and insertion.  
S3045 OVERVIEW  
The S3045 byte interleave chip implements SONET/  
SDH byte interleave functions required to multiplex/  
demultiplex four STS-12/STM-4 data streams into a  
single STS-48/STM-16 data stream. Each of the four  
STS-12/STM-4 transmit/receive data streams uses  
an 8-bit parallel interface with parity to maintain com-  
patibility with industry standard network interface  
processors. The STS-48/STM-16 data stream uses  
an 8-bit parallel LVDS data path to be compatible  
with the S3041/S3042 Mux/Demux chipset. The  
block diagram in Figure 2 shows the basic operation  
of the chip. This chip can be used with the S3041  
and S3042 to implement the front end of SONET  
equipment. The chip includes byte interleave cir-  
cuitry along with B1 calculation, M1 calculation, J0/  
Z0 insertion, and B1 verification circuitry. STS-48/  
STM-16 data stream is monitored in the receive path  
for OOF and LOS, and alarm outputs are generated.  
9. STS-48/STM-16 compatible 8-bit wide 311 MHz  
LVDS output (311DATOUT).  
Receiver Operations  
1. STS-48/STM-16 compatible 8-bit wide 311 MHz  
LVDS input (311DATIN).  
2. OOF and LOS states are monitored and alarms  
are generated.  
3. B1 extraction and calculation of the STS-48/STM-16  
frame.  
4. B1 calculation of the number one STS-12/STM-4  
frame.  
5. Frame synchronous descrambling.  
6. B1 compare and error indication (B1ERR) genera-  
tion for the STS-48/STM-16 frame.  
The S3045 is divided into a transmitter section and a  
receiver section. The sequence of operation is as  
follows:  
7. Insert the number one STS-12/STM-4 B1 parity byte  
into the number one STS-12/STM-4 frame and insert  
errors if any found in the STS-48/STM-16 B1 parity  
byte.  
Transmitter Operations  
1. 32-bit LVTTL parallel input from four 8-bit STS-  
12/ STM-4 data streams (PIN A,B,C,D) with parity  
(PARIN A,B,C,D).  
8. 32-bit (4 x 8 bit) parallel output of four STS-12/  
STM-4 data streams (POUT A,B,C,D) with parity  
output (PAROUT A,B,C,D).  
2. Four Byte interleave conversion Mux.  
3. Section-trace insertion (J0/Z0).  
Suggested Interface Devices  
4. M1 calculation (addition of four STS-12/STM-4  
M1 values) and insertion into the number one  
STS-12/STM-4 location.  
AMCC  
AMCC  
AMCC  
AMCC  
AMCC  
S3040/S3047  
S3041  
Clock Recovery Device  
OC-48 Mux  
5. M1 insertion of zero into STS-12/STM-4 number  
two, three, and four locations.  
S3042  
OC-48 Demux  
CONGO S1201 POS/ATM SONET Mapper  
NILE S1202 ATM SONET Mapper  
6. Four B2 parity byte calculations and insertions for  
STS-1 frames after M1 insertions.  
S3045 Acronym List  
SDH  
SEF  
Synchronous Digital Heirarchy  
Severely Errored Frame  
BER  
CDR  
LOS  
OOF  
ORX  
OTX  
Bit Error rate  
Clock and Data Recovery  
Loss Of Signal  
SONET –  
Synchronous Optical Network  
Synchronous Transport Module  
Synchronous Transport Signal  
STM  
STS  
Out Of Frame  
Optical Receiver\  
Optical Transmitter  
2
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Figure 2. Functional Block Diagram  
*A. J0/Z0 Insertions  
*B. M1 Calculations &  
insertions  
*C. B2 Calculations  
and Insertions  
*D. B1 Calculation  
and Insertion  
PARFPTXSEL  
SCRBENB  
FIFO  
Register  
Parity  
B2/M1SELB  
B1SELB  
1
1
Calculation  
PARIN A  
PINA  
PARERR A  
8
PICLK  
8 bit  
Register  
PARIN B  
PINB  
PARERR B  
8
8
8
D*  
A* B*  
8
C*  
Scrambler  
311DATOUT  
Register  
Register  
Register  
Byte  
Interleave  
Mux  
PICLK  
1
1
PARIN C  
PINC  
PARERR C  
8
8
B1 Parity  
Calculation &  
Insertion  
311CLKOUT  
J0/Z0SEL  
PICLK  
PARIN D  
PIND  
PARERR D  
READP/N  
PICLK  
Register  
PULSEP/N  
PCLK A,B,C,D  
Transmit Clock  
Divider  
311TCLK  
Frame  
Control  
Counter  
4
TIFP A,B,C,D  
SYNCRSTB  
TRANSMITTER  
RECEIVER  
LB_FRAME  
PARSEL  
DSCRBENB  
PAROUT A,B,C,D  
POUT A  
SDVBB  
SDLVPECL  
SDLVTTL  
SQUELCHB  
DLEB  
4
8
8
MUX  
2*  
Byte  
Interleave  
Demux  
1*  
POUT B  
Register  
8
8
8
Descrambler  
POUT C  
POUT D  
311DATIN  
PARFPRXSEL  
POCLK A,B,C,D  
4
311CLKIN  
FRAME  
PULSEOOF  
OOF  
Frame  
Control  
Counter &  
Alarm  
4
B1 Parity  
FP A,B,C,D  
STS-12  
Calculation  
Insertion  
Calculation for  
STS-48 and the  
number 1 STS-12  
frames  
Detection  
J0FP  
FPSEL  
LOS  
*1. Extract STS-48 B1  
for Comparison  
Receive Clock  
Divider  
*2. LOS monitoring  
B1ERR  
RSTB  
3
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
the byte following the last byte of the first row (last  
C1 byte) of the STS-12/STM-4 section overhead.  
This bit and all subsequent bits to be scrambled will  
be added modulo 2 to the output from the X7 position  
of the scrambler (A1, A2, or C1 bytes are not  
scrambled). The scrambler will run continuously  
throughout the complete STS-48/STM-16 frame. A  
set of signals from the frame control counter block  
controls when the scrambler is on, off, or reset.  
ARCHITECTURE/FUNCTIONAL DESIGN  
Transmit Operation  
The S3045 transmit section performs the byte inter-  
leaving stage in the processing of a transmit  
SONET/SDH STS-48/STM-16 byte wide data  
stream. It converts four byte wide STS-12/STM-4  
data streams into a single byte serial STS-48/STM-  
16 data stream. The byte interleaved parity (B1) is  
calculated over the entire STS-48/STM-16 frame and  
inserted into the appropriate B1 location. In each  
STS-48/STM-16 frame there is one B1 byte located  
in the first STS-1 frame. The M1 byte is calculated  
(addition of four STS-12/STM-4 M1 bytes) and in-  
serted into the number one STS-12/STM-4 M1  
position. Zero is inserted into the number two, three,  
and four STS-12/STM-4 M1 positions. The section-  
trace bytes (J0/Z0) can be optionally inserted by  
setting the J0/Z0SEL high. The SONET/SDH scram-  
bler can be enabled or disabled by the SCRBENB  
input.  
Frame Synchronization  
The four STS-12/STM-4 input data streams from the  
four controllers must be frame aligned before this  
data is fed into the S3045 since the S3045 does not  
have any data buffering. The four controllers each  
output an active high framing position signal that is  
input into the TIFP inputs of the S3045 that marks  
the frame alignment on the output bus. This signal  
goes high for a single 77.76 MHz clock period during  
the first synchronous payload envelope byte immedi-  
ately following the C1 bytes. The TIFP A,B,C,D  
inputs associated with each of the four STS-12/STM-  
4 data streams must be high at the same time so as  
to indicate frame alignment of all four data streams.  
These frame pulses will indicate frame alignment  
with the first payload byte of the STS-12/STM-4  
frames. When frame alignment occurs, valid data will  
start to be output by the next valid frame. Valid B1  
and B2 parity bytes will be output on the following  
frame. Otherwise random data will be clocked out of  
the 311DATOUT[7:0] output. In the event that all  
four pulses are not high at the same time (frames  
are not exactly aligned), a reset sequence will be  
generated by the S3045 to re-synchronize the four  
data streams.  
STS-48/STM-16 Byte Interleave Multiplexing  
The byte interleave mux shown in Figure 2 takes in  
the four byte wide STS-12/STM-4 data streams four  
bytes at a time and outputs byte wide STS-48/STM-  
16 data stream. The byte interleave mux inputs are  
registered on the STS-12/STM-4 interface. The mux  
first takes four bytes from the A input, followed by  
four from B, four from C, and four from D. The pat-  
tern is repeated as data on the A, B, C, and D inputs  
are registered and passed into a four word deep  
pipeline register. Each of the four data words are  
then latched into holding registers. A counter at the  
STS-48/STM-16 byte rate (311 MHz) controls a mux  
that loads data into a register at the STS-48/STM-16  
byte rate and is then transmitted. STS-48/STM-16  
byte interleaving must be done four bytes at a time,  
where as STS-3/STM-1 and STS-12/STM-4 is ac-  
complished one byte at a time.  
In order to guarantee the synchronization of the four  
controllers, a reset sequence will be applied by the  
S3045. This reset will align the four STS-12/STM-4  
inputs for multiplexing. Note that parity errors may  
be erroneously generated during this reset se-  
quence. Figure 3 depicts the following synchroniza-  
tion sequence.  
Scrambler  
The scrambler can be utilized in order to guarantee  
a suitable bit pattern, which prevents a long se-  
quence of 1’s or 0’s. The frame synchronous  
scrambler can be optionally used to scramble the  
STS-48/STM-16 data stream. The SONET scram-  
bling generator polynomial of 1 + x6 + x7 with a  
sequence length of 127 is used. The scrambler will  
be reset to “1111111” on the most significant bit of  
1. The SYNCRSTB will be asserted low to reset the  
four controllers upon a misalignment of the four  
TIFP pulses and will be held low for a minimum  
16 PCLK A,B,C,D clock cycles (77.76 MHz).  
2. The transmitter data clock (PCLK A,B,C,D) will  
stop for a minimum 16 clock cycles while the  
SYNCRSTB is asserted low.  
4
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
3. The SYNCRSTB will be removed (de-asserted  
high) from the four controllers for a minimum 16  
PCLK A,B,C,D clock cycles.  
Clock Generation  
The clock generation circuitry generates the 77.76  
MHz PCLK A,B,C,D clock required for the byte wide  
STS-12/STM-4 interface from the 311 MHz clock  
(311TCLK) required for the byte wide STS-48/STM-  
16 interface.  
4. The transmit data clock (PCLK A,B,C,D) will con-  
tinue to be stopped for a minimum 16 clock  
cycles after reset is de-asserted and then start  
clocking.  
Bit Interleaved Parity - N  
5. STS-12/STM-4 data out of the four controllers will  
start to flow into the S3045. The S3045 will  
search for a high pulse on the TIFP input pulses  
on all four channels. If the four controllers are not  
frame aligned in 250us this sequence will restart.  
Bit Interleaved Parity-N (BIP-N) is a method of error  
monitoring. Even parity is used for the BIP-8 calcula-  
tion. The transmitting equipment generates an N-bit  
code over a specified portion of the signal in such a  
manner that the first bit of the code provides even  
parity over the first bit of all N-bit sequences on the  
covered portion of the signal, the second bit provides  
even parity over the second bit of all N-bit se-  
quences within the specified portion, etc. Even parity  
is generated by setting the BIP-N bits so that there  
are an even number of ones in each of all N-bit  
sequences including the BIP-N.  
Frame Control Counter  
The frame counter receives the TIFP signals indicat-  
ing the frame boundaries and counts 38,880 bytes (9  
rows x 90 columns x 48 STS-1) to ensure that the  
transmitter is receiving synchronous TIFP pulses.  
The frame counter keeps track of the overhead bytes  
so that proper location and insertion of bytes is ac-  
complished.  
B1 Parity Calculation  
The frame control block outputs a frame synchro-  
nous reset signal (SYNCRSTB) that frame aligns the  
STS-12/STM-4 network interface processors. The  
Transmit Input Frame Pulse (TIFP) signal indicates  
to the frame counter the frame position of the input  
data channel. If the transmitter does not get four  
simultaneous TIFP signals after 250us the  
SYNCRSTB pin is asserted low. The frame counter  
controls the output mux on the scrambler to allow  
scrambling, parity byte (B1) generation and inser-  
tion, M1 calculations and insertions, Z0 calculations  
insertions, and parity byte (B2) generations and in-  
sertions.  
This byte is allocated for regeneration section error  
monitoring. This byte will be calculated using even  
parity. Even parity is generated by setting the bit  
interleaved parity bits so that there is an even num-  
ber of ones in each monitored partition of the signal.  
The interleaved even parity byte B1 is calculated  
over the entire scrambled STS-48/STM-16 frame  
and inserted into the B1 location of the next frame  
before going through the scrambling process. The  
computed bit interleaved parity is only placed in the  
B1 byte of the first STS-1 signal of the STS-48/STM-  
16 frame before scrambling (one B1 byte is valid in a  
Figure 3. S3045 Synchronous Reset Functional Timing Diagram  
PCLK A,B,C,D  
16 cycles  
16 cycles  
TIFP A  
TIFP B  
TIFP C  
TIFP D  
TIFP D misaligned  
SYNCRSTB  
5
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
single STS-48/STM-16 frame). The B1 parity byte is  
calculated after the M1 bytes are inserted, the sec-  
tion-trace J0/Z0 bytes are inserted, and the B2 parity  
bytes are inserted.  
frame). These parity bytes are optionally calculated  
and inserted after the M1 bytes are inserted into the  
STS-48/STM-16 and before the B1 is calculated and  
inserted. The B2/M1SELB select line allows for B2  
and M1 calculations to be turned off.  
B2 Parity Calculation  
M1 Byte ( B2 Parity Error Count )  
The B2 byte is allocated in each STS-1 frame for a  
line error monitoring function. The interleave even  
parity byte B2 is optionally calculated and inserted  
only over the four STS-1 frames that the four M1  
bytes were inserted (the B2 byte will not be calculated  
for all 48 STS-1’s). These four B2 bytes are calcu-  
lated over the STS-1 frames of the previous STS-48/  
STM-16 frame except for the first three rows of the  
section overhead (computed over all bits of the line  
overhead and capacity envelope of the previous STS-  
1 frame before scrambling) and is inserted into the  
appropriate B2 locations (locations of the B2 bytes  
are found by control from the frame counter block) of  
the current frame before scrambling. These bytes will  
be inserted within a STS-48/STM-16 signal (48 B2  
bytes are contained in a single STS-48/STM-16  
The STS-48/STM-16 M1 byte is calculated by ex-  
tracting each of the four M1 bytes of the STS-12/  
STM-4 frames and adding them together and insert-  
ing this new M1 byte into the STS-48/STM-16 M1  
byte. For STS-48/STM-16 rate the M1 count will  
truncate at 255 (never report more than 255 errors).  
The STS-48/STM-16 M1 byte will only be transmitted  
over the number one STS-12/STM-4 data stream.  
The other 3 STS-12/STM-4 M1 bytes will be inserted  
with the value of zero.  
Section-trace Insertion (J0/Z0) Bytes  
The section-trace bytes (J0/Z0) can optionally be  
filled by setting J0/Z0SEL as indicated below in  
Table 1.  
Table 1. Section-Trace Insertion (J0/Z0) Bytes  
J0/Z0SEL  
Description  
0
Transparent Operation - J0/Z0 bytes are passed through with no modification.  
Byte 1 of 48 (J0 byte) is passed through with no modification (transparent) and  
bytes 2 through 48 (Z0 bytes) are filled with the values of 02hex to 30hex (48  
decimal) respectively.  
1
6
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
RECEIVER OPERATION  
S3045  
Receive Frame Counter  
The S3045 byte interleave receive section converts  
the byte wide STS-48/STM-16 LVDS data stream  
into four byte wide STS-12/STM-4 LVTTL data  
streams with parity. The B1 parity byte is calculated  
over the STS-48/STM-16 frame and over the number  
one STS-12/STM-4 frame. The STS-48/STM-16 B1  
parity byte is compared to the one received, if an  
error exists then B1ERR will be asserted and the  
STS-12/STM-4 parity B1 will be inserted with the  
same number of bit errors found in the STS-48/STM-  
16 B1 parity byte. Also, if no errors exist then the  
calculated STS-12/STM-4 B1 will be inserted into the  
number one STS-12/STM-4 frame. A frame synchro-  
nous descrambler can be optionally disabled by the  
DSCRBENB input. The input data stream is moni-  
tored for Loss of Signal (LOS) and Out of Frame  
(OOF) and alarms are generated for each one.  
The frame counter receives the FRAME signal indi-  
cating the frame boundaries and counts 38,880  
bytes (9 rows x 90 columns x 48 STS-1) to ensure  
that the receiver is receiving synchronous FRAME  
pulses. The frame counter keeps track of the over-  
head bytes so that proper location and insertion of  
bytes is accomplished. In diagnostic loopback mode,  
a loopback frame (LB_FRAME) signal is generated  
internally and the FRAME input is disabled.  
Clock Generation  
The clock generation circuitry generates the 77.76  
MHz POCLK clock required for the byte wide STS-12/  
STM-4 interface from the 311 MHz clock (311CLKIN)  
required for the byte wide STS-48/STM-16 interface.  
B1 Parity Calculation and Compare  
The B1 byte is allocated for regeneration section error  
monitoring. This byte will be calculated using even  
parity. The section bit interleaved parity (BIP-8) error  
detection code B1 will be optionally calculated for ev-  
ery STS-48/STM-16 frame before descrambling and  
for the number one STS-12/STM-4 frame after  
descrambling. The B1 value is compared to the ex-  
tracted value of the STS-48/STM-16 B1 parity byte  
after descrambling in the following frame. B1 errors  
will be shown at the B1ERR output when the B1SELB  
is active. The calculated STS-12/STM-4 B1 parity  
byte will be inserted after descrambling into the num-  
ber one STS-12/STM-4 frame if there are no errors  
found on the STS-48/STM-16 B1 parity byte. If there  
are errors found with the STS-48/STM-16 B1 parity  
byte, the number of bit errors (1 to 8) will be passed  
onto the STS-12/STM-4 B1 parity byte for insertion  
into the number one STS-12/STM-4 frame. The num-  
ber one STS-12/STM-4 frame is output on the  
POUT[7:0]A data bus. The following functional timing  
diagram depicts the B1ERR timing.  
Descrambling  
The byte wide STS-48/STM-16 data stream is op-  
tionally descrambled using the SONET frame syn-  
chronous descrambler with a generator polynomial  
of 1 + x6 + x7 with a sequence length of 127. The  
descrambler algorithm is identical to the scrambler  
algorithm. The descrambler will be reset to  
“1111111” on the most significant bit of the byte fol-  
lowing the last byte of the first row of the STS-12/  
STM-4 section overhead. This bit and all subsequent  
bits to be descrambled will be added modulo 2 to the  
output from the X7 position of the descrambler (A1,  
A2, or C1 bytes are not descrambled). The  
descrambler will run continuously throughout the  
complete STS-48/STM-16 frame. A signal from the  
frame counter block controls when the descrambler  
is on, off, or reset.  
Figure 4. B1 Error (B1ERR) Functional Timing Diagram  
POCLK  
B1ERR  
low for a minimum of 2  
clock cycles  
7
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Loss of Signal (LOS)  
FRAMER (In-frame, going out-of-frame)  
The frame acquisition algorithm determines whether  
the receiver is in-frame or out-of-frame. In-frame is  
defined as the state where the frame boundaries are  
known. Out-of-Frame (OOF) is defined as the state  
where the frame boundaries of the incoming signal  
are unknown. OOF is also referred to as the Se-  
verely Errored Frame (SEF) in the SONET  
standards. The frame pulse (FRAME) input of the  
S3045 indicates if the frame boundaries are known  
(in-frame) or unknown (out-of frame).  
The Loss of Signal (LOS) block monitors the incom-  
ing data stream for all zero’s. When a minimum of 27  
µsec of all zero’s are detected, a loss of signal (LOS)  
is declared (high signal). LOS is deactivated (low  
signal) when two consecutive valid framing patterns  
are detected and no LOS condition is detected in  
between. LOS is updated on the falling edge of  
POCLKA,B,C,D.  
J0 Frame Pulse  
The J0 Frame Pulse output will be active high when  
the J0 byte is presented on the POUT[7:0]A data  
bus. Figure 6 depicts the functional timing of this  
signal.  
An Out-of-Frame (OOF) also known as an SEF con-  
dition on an STS-48/STM-16 signal will be declared  
when a minimum of four consecutive errored framing  
patterns have been received. The maximum SEF de-  
tection time will be 625 µs for a random signal. The  
framing algorithm used to check the alignment is  
such that 10-3 BER does not cause an SEF more  
than once every 6 minutes. This algorithm examines  
the 48th A1 (F6h) byte and the first four bits of the  
first A2 (28h) byte for a total of 12 bits to guarantee  
this requirement.  
FP Frame Pulse  
The FP output indicates frame boundaries in the in-  
coming data stream. FP pulses high for one POCLK  
cycle when the third A2 byte of the framing se-  
quence is valid on the POUT[7:0]A data bus. Figure  
7 depicts the functional timing of this signal.  
STS-48/STM-16 Byte Interleave Demux  
When in an SEF condition the S3045 will assume  
the frame has been recovered and declare an in-  
frame condition on detecting two successive error-  
free framing patterns. This implementation of the  
frame recovery circuit achieves realignment follow-  
ing a declared SEF within 250-µs.  
The byte interleave demux shown in Figure 2 con-  
verts the byte wide STS-48/STM-16 data stream into  
four byte wide STS-12/STM-4 data streams. The  
data is byte de-interleaved using four bytes at a  
time. The data is output with a 77.76 MHz clock.  
Figure 5. OOF State Machine  
4 consecutive errored framing patterns  
Reset  
OOF (SEF)  
STATE  
NORMAL STATE  
2 consecutive error free  
framing patterns  
8
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
It is important to note that in squelch mode there  
may be up to 3.2 nsec shortening or lengthening of  
the POCLK A,B,C,D cycle, resulting in an apparent  
phase shift in the POCLK at the de-assertion of the  
SD condition. Another similar phase shift will occur  
when the SD condition is reasserted. Figure 8 de-  
picts this operation.  
OTHER OPERATING MODES  
Diagnostic Loopback  
When the Diagnostic Loopback Enable (DLEB) input is  
low, a loopback from the transmitter to the receiver at  
the serial data rate can be set up for diagnostic pur-  
poses. In loopback mode the STS-48/STM-16  
transmitter outputs (311DATOUT[7:0], 311CLKOUT  
and an internally generated Frame pulse on the third  
A2 byte (that is not accessible to the outside of the  
chip)) are internally connected to the STS-48/STM-16  
receiver inputs (311DATIN[7:0], 311CLKIN, and  
FRAME). In loopback mode the STS-48/STM-16 trans-  
mitter outputs (311DATOUT[7:0] and 311CLKOUT) are  
still active.  
In the normal operating mode with SQUELCHB input  
inactive (high), there will be no phase discontinuities  
at the POCLK A,B,C,D output during signal loss or  
re-acquisition (assuming operation with continuous  
clock from the CRU device such as the AMCC  
S3040 or S3047). Figure 9 depicts this operation.  
Byte wide Parity Calculation  
Odd parity or even parity can be calculated using the  
PIN A,B,C,D and POUT A,B,C,D data busses by set-  
ting the PARSEL line high for even parity or low for  
odd parity. Even parity is generated by setting the  
parity bit so that there are an even number of ones  
throughout the 9 bits. Odd parity is generated by  
setting the parity bit so that there is an odd number  
of ones throughout the 9 bits.  
“Squelched Clock” Operation  
Some integrated optical receiver/clock recovery  
modules force their recovered serial receive clock  
output to the logic zero state (squelched clock) if the  
optical signal is removed or reduced below a fixed  
threshold. This condition is accompanied by the ex-  
pected deassertion of the Signal Detect (SD) output.  
Reset Operation  
The S3045 has been designed for operation with  
clock recovery devices that provide continuous serial  
clock for seamless down stream clocking in the  
event of optical signal loss. For operation with an  
optical transceiver that provides the “squelched  
clock” behavior as described above, the S3045 can  
be operated in the “squelched clock mode” using the  
SQUELCHB input.  
The RESET (RSTB) input forces all the internal logic  
of the S3045 to its deasserted state. All of the se-  
quential logic inside the S3045 will remain static for  
as long as reset is asserted. The RESET input is  
implemented using a schmitt type receiver. The RE-  
SET input is asynchronous to the input clock.  
RESET should be asserted for at least 30 nano sec-  
onds after power up for proper operation. During  
reset PCLK A,B,C,D and POCLK A,B,C,D will stop,  
and SYNCRSTB will remain high.  
In squelch mode, the 311CLKIN is used for all re-  
ceiver timing when the SDLVPECL or SDLVTTL  
inputs are in the active state. (SDLVPECL and  
SDLVTTL are in opposite logical states.) When the  
SDLVPECL or SDLVTTL inputs are placed in the  
inactive state (usually by the de-assertion of the Sig-  
nal Detect [SDLVPECL and SDLVTTL are in the  
same logical state] from the optical transceiver/clock  
recovery unit) the transmitter serial clock (311TCLK)  
will be used to maintain timing in the receiver sec-  
tion. This will allow the POCLK A,B,C,D to continue  
to run and the parallel outputs to flush out the last  
received characters and assume the all zero state  
imposed at the serial data input.  
9
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Figure 6. J0 Frame Pulse Functional Timing Diagram  
J0FP  
POUTA[7:0]  
J0  
Z0  
Z0  
Z0  
Z0  
Z0  
POUTB[7:0]  
POUTC[7:0]  
POUTD[7:0]  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Z0  
Figure 7. FP Frame Pulse Functional Timing Diagram  
FP  
POUTA[7:0]  
A1  
A2  
A2  
A2  
A2  
A2  
POUTB[7:0]  
POUTC[7:0]  
POUTD[7:0]  
A1  
A1  
A1  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
A2  
10  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Figure 8. Functional Operation of SQUELCH Mode  
SDLVPECL  
Note: POCLK A,B,C,D is now  
derived from 311TCLKP/N  
(from the transmitter)  
POCLK A,B,C,D  
SDLVTTL = LOW  
SQUELCHB = LOW  
Note: A shortening or lengthening of  
POCLK A,B,C,D of up to 3.2 ns.  
(Assuming 311CLKIN is within  
its specification)  
Figure 9. Functional Operation of Non-SQUELCH Mode  
SDLVPECL  
POCLK A,B,C,D  
SDLVTTL = LOW  
SQUELCHB = HIGH  
Note: POCLK A,B,C,D continues  
to be derived from 311CLKINP/N  
(from the receiver)  
11  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Table 2. Transmitter Input Pin Assignment and Descriptions  
Pin Name  
PICLK  
Level  
I/O  
Pin #  
Description  
LVTTL  
I
129  
Parallel Input Clock. PICLK is a 77.76 MHz input clock which  
PIN[7:0] is aligned. PICLK is used to transfer the data on the PIN  
inputs into a holding register. The rising edge of PICLK samples  
PIN A, B, C, D[7:0].  
PINA[7:0]  
PINB[7:0]  
PINC[7:0]  
PIND[7:0]  
LVTTL  
LVTTL  
I
I
180–173, Parallel Data Input. A 77.76 Mbytes/sec word, aligned to the  
156–149, PICLK parallel input clock. PIN[7] is the most significant bit  
121–117, (corresponding to bit 1 of each word, the first bit transmitted).  
115–113, PIN[0] is the least significant bit (corresponding to bit 8 of each  
94–91,  
89–86  
word, the last bit transmitted). PIN[7:0] is sampled on the rising  
edge of PICLK.  
SCRBENB  
198  
Scramble Enable. Active low. When active, the frame  
synchronous scrambler is enabled. When inactive, the scrambler  
is disabled. This signal is static and must not be changed in normal  
operation.  
311TCLKN  
311TCLKP  
LVDS  
I
I
30  
31  
311 MHz Transmit Clock. Used by the transmitter to generate the  
77.76 MHz clocks and retime the STS-48/STM-16 byte wide data.  
TIFPA  
TIFPB  
TIFPC  
TIFPD  
LVTTL  
172  
148  
112  
85  
Transmit Input Frame Pulse. Active high. When active, it indicates  
the frame position of the transmit data (PIN[7:0]). TIFP goes high  
for a single PCLK A, B, C, D (77.76 MHz) period during the first  
synchronous payload envelope byte after 12 C1 bytes. TIFP is  
clocked in on the rising edge of PICLK.  
PARINA  
PARINB  
PARINC  
PARIND  
LVTTL  
LVTTL  
I
I
181  
146  
111  
84  
Parity Input. Odd or even parity depending on the input of parity  
select (PARSEL) for the 8 bit PIN[7:0] A, B, C, D data bus. PARIN  
is clocked in on the rising edge of PICLK.  
B2/M1SELB  
200  
B2/M1 Parity Byte and Parity Count Select. Active low. When  
inactive the B2/M1 byte calculations and insertions are disabled.  
When active, normal operation occurs (B2 and M1 calculations  
and insertions are enabled). This signal is static and must not be  
changed in normal operation.  
J0/Z0SEL  
LVTTL  
I
I
205  
Section-Trace Insertion Select. Select pin, select section-trace  
bytes J0/Z0 options. When low the J0/Z0 bytes are passed  
through with no modification. When high, byte 1 of 48 (J0 byte) is  
passed though with no modification (transparent) and bytes 2  
through 48 (Z0 bytes) are filled with the values of 02hex to 30hex  
(48 decimal) respectively. (See Table 1.) This signal is static and  
must not be changed in normal operation.  
PULSEP  
PULSEN  
LVDS  
52  
51  
PULSE Input. This input is used to generate the READP/N output  
by synchronizing the PULSEP/N input to the 311TCLK input clock  
through a register. (Required for operation with the AMCC S3041  
OC-48 TX Mux).  
PARFPTXSEL  
LVTTL  
208  
Parity Frame Pulse Transmit Select. When low, parity is calculated  
over the data bus PIN[7:0] A, B, C, D. When high, parity is  
calculated over the PIN[7:0] A, B, C, D data bus and the Transmit  
Input Frame Pulse (TIFP A, B, C, D). This signal is static and must  
not be changed in normal operation.  
12  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Table 3. Transmitter Output Pin Assignment and Descriptions  
Pin Name  
Level  
I/O  
Pin #  
Description  
PCLKA  
LVTTL  
O
128  
127  
126  
125  
Parallel Clock. A 77.76 MHz clock generated by dividing the  
internal 311TCLK by four. It is normally used to coordinate byte  
wide transfers between the STS-12/STM-4 overhead processors  
and the S3045 device.  
PCLKB  
PCLKC  
PCLKD  
311DATOUT-  
P/N[7:0]  
LVDS  
O
48–41  
39–32  
311 Mbit Data. STS-48/STM-16 byte wide data path. A 311  
Mbytes/sec word, aligned to the 311 MHz parallel output clock  
(311CLKOUT). 311DATOUT [7] is the most significant bit  
(corresponding to bit 1 of each word, the first bit transmitted).  
311DATOUT[0] is the least significant bit (corresponding to bit 8  
of each word, the last bit transmitted).  
311CLKOUTP  
311CLKOUTN  
LVDS  
O
O
27  
26  
311 MHz Clock. 311 CLKOUT is a clock for the transmit STS-  
48/STM-16 byte wide data path. A 311 MHz output clock, which  
311DATAOUT [7:0] is aligned.  
SYNCRSTB  
LVTTL  
122  
Synchronous Reset. Active low. When active, the Network  
Interface Processors are reset to synchronize the four STS-  
12/STM-4 incoming data streams. Figure 3 shows the S3045  
synchronous reset timing diagram.  
PARERRA  
PARERRB  
PARERRC  
PARERRD  
LVTTL  
O
182  
145  
110  
83  
Parity Error Output. Active high. Indicates to the controller that a  
parity error has been detected on the PIN[7:0] A, B, C, D data bus  
on a previous byte of data. When active, a parity error has been  
received. When inactive, PIN[7:0] A, B, C, D data has been  
received without parity errors. PARERR output can be delayed by  
one to five PCLK cycles due to the internal data FIFO. See Figure  
10. Note that during a sync reset condition and a hardware reset,  
parity errors may erroneously be generated.  
READP  
READN  
LVDS  
O
54  
53  
Read Output. This output is the result of synchronizing the  
PULSEP/N input to the 311TCLK input clock through a register.  
(Required for operation with the AMCC S3041.)  
13  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Table 4. Receiver Input Pin Assignment and Descriptions  
Pin Name  
Level  
I/O  
Pin #  
Description  
311DATINN/P  
[7:0]  
LVDS  
I
21–14,  
11–4  
311 Mbit Data Inputs. Parallel STS-48/STM-16 data bus, a 311  
Mbyte/sec word aligned to the parallel input [311CLKIN]. 311DATAIN  
[7] is the most significant bit (corresponding to bit 1 of each word, the  
first bit received). 311DATAIN [0] is the least significant bit  
(corresponding to bit 8 of each word, the last bit received).  
311DATAIN [7:0] is latched on the rising edge of 311CLKIN.  
311CLKINN  
311CLKINP  
LVDS  
LVDS  
I
I
25  
24  
311 MHz Clock. 311CLKIN is a 311 MHz byte rate input clock that  
is aligned to 311DATAIN [7:0] byte serial input data. 311DATIN [7:0]  
and FRAME are clocked in on the rising edge of the 311CLKIN.  
FRAMEN  
FRAMEP  
2
1
Frame. Active high. When active, it indicates that the third A2 byte  
of the framing sequence is valid on the 311DATIN<7:0> pins.  
Therefore it indicates frame boundaries in the incoming data stream  
(311DATIN[7:0]).  
DSCRBENB  
SQUELCHB  
LVTTL  
LVTTL  
I
I
199  
201  
Descrambler Enable. Active low. When active, the frame  
synchronous descrambler is enabled. When inactive the frame  
synchronous descrambler is disabled. This signal is static and must  
not be changed in normal operation.  
Squelch Clock Mode. Active low. Set inactive when a clock recovery  
device used provides a continuous clock during signal loss or re-  
acquisition. Set active when the clock recovery device used does  
not provide a continuous clock during signal loss or signal  
acquisition. When active and SDLVPECL/SDLVTTL is inactive  
(SDLVPECL and SDLVTTL are in the same logical states) the  
transmitter serial clock (311TCLK) will be used to maintain timing  
in the receiving section. This signal is static and must not be  
changed in normal operation.  
SDLVTTL  
LVTTL  
I
192  
Signal Detect. Active High when SDLVPECL is tied to logic 0. Active  
Low when SDLVPECL is held at logic 1. A single-ended LVTTL  
input to be driven by the external optical receiver module to indicate  
a loss of received optical power. When SDLVTTL is inactive, the  
data on the 311DATIN[7:0] pins will be internally forced to a  
constant zero with the descrambler bypassed. Note: When B1SEL  
is active, a B1 byte is inserted into each frame. When SDLVTTL is  
active, data on the 311DATIN[7:0] pins will be processed normally.  
SDLVPECL  
LVPECL  
I
191  
Signal Detect. Active High when SDLVTTL is held at logic 0. Active  
Low when SDLVTTL is held at logic 1. A single-ended LVPECL  
input to be driven by the external optical receiver module to indicate  
a loss of received optical power. When SDLVPECL is inactive, the  
data on the Serial Data in 311DATIN[7:0] pins will be internally  
forced to a constant zero with the descrambler bypassed. Note:  
When B1SEL is active, a B1 byte is inserted into each frame. When  
SDLVPECL is active, data on the 311DATIN[7:0] pins will be  
processed normally. When SDLVTTL is to be connected to the  
optical receiver module instead of SDLVPECL, then SDLVPECL  
should be tied High to implement an active low Signal Detect.  
14  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Table 4. Receiver Input Pin Assignment and Descriptions (Continued)  
Pin Name  
SDVBB  
Level  
I/O  
Pin #  
Description  
I
190  
Signal Detect. Bias pin for the SDLVPECL input. Biased internally  
to VDD -1.3V.  
FPSEL  
LVTTL  
I
I
204  
207  
Frame Pulse Select. When low the FRAME input is used to  
generate the FP A, B, C, D pulse when the third A2 byte is output.  
When high, the FP A, B, C, D output is internally generated using  
the A1A2 frame boundary. The FP A, B, C, D is asserted high when  
the third A2 (28h) byte is output. For normal operation set high. This  
signal is static and must not be changed in normal operation.  
PARFPRXSEL  
LVTTL  
Parity Frame Pulse Receive Select. When low, parity is calculated  
over the data POUT[7:0] A, B, C, D. When high, parity is calculated  
over the data POUT[7:0] A, B, C, D and the frame pulse (FP A, B,  
C, D) output. This signal is static and must not be changed in normal  
operation.  
Table 5. Receiver Output Pin Assignment and Descriptions  
Pin Name  
Level  
I/O  
Pin #  
Description  
POUTA[7:0]  
POUTB[7:0]  
POUTC[7:0]  
POUTD[7:0]  
LVTTL  
O
168–161 Parallel Data Output. Parallel data bus, a 77.76 Mbyte/sec word,  
140–133 aligned to the parallel output clock (POCLK). POUT[7] is the most  
106–99 significant bit (corresponding to bit 1 of each word, the first bit  
79–72  
received). POUT[0] is the least significant bit (corresponding to bit  
8 of each word, the last bit received). POUT[7:0], LOS, PAROUT  
A, B, C, D, B1ERR, FP A, B, C, D and OOF are updated on the  
falling edge of POCLK.  
POCLKA  
POCLKB  
POCLKC  
POCLKD  
LVTTL  
LVTTL  
LVTTL  
O
O
O
169  
141  
107  
80  
Parallel Output Clock. POCLKA, B, C, D are 77.76 MHz byte rate  
output clocks that are aligned to POUT[7:0]A, B, C, D byte serial  
output data. POUT[7:0] is updated on the falling edge of POCLK.  
FPA  
FPB  
FPC  
FPD  
157  
130  
95  
Frame Pulse. Indicate frame boundaries in the incoming data  
stream. FP pulses high for one POCLK cycle when the third A2 byte  
of the framing sequence is valid on the POUTA[7:0] data. FP is  
updated on the falling edge of POCLK.  
69  
OOF  
185  
Out of Frame. The Out of Frame (OOF) signal is active when the  
S3045 has detected an out of frame condition. The OOF is inactive  
when the S3045 is in frame. An OOF declaration occurs when four  
consecutive errored framing patterns are received. OOF is used to  
enable upstream framing pattern detector to search for the framing  
pattern. Figure 13 depicts the functional timing of this signal.  
15  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Table 5. Receiver Output Pin Assignment and Descriptions (Continued)  
Pin Name  
Level  
I/O  
Pin #  
Description  
PULSEOOF  
LVTTL  
O
186  
Pulse Out of Frame. Active high. The Out of Frame (OOF) signal  
will pulse for a minimum of 7.5ns when the S3045 is out of frame  
and when the A1A2 boundary is not received for 250µs. The  
PULSEOOF is inactive when the S3045 is in frame. An OOF  
declaration occurs when four consecutive errored framing patterns  
are received. OOF is used to enable upstream framing pattern  
detector to search for the framing pattern. Figure 13 depicts the  
functional timing of this signal.  
J0FP  
LVTTL  
LVTTL  
O
O
206  
183  
J0 Frame Pulse. Active high. Active when the J0 byte is presented  
on the POUTA[7:0] data bus.  
B1ERR  
B1 Parity Error. Active high. Indicates that a B1 bit error has been  
detected when high. For each frame B1ERR will pulse a maximum  
of 8 times. In between pulses (bit errors) the B1ERR will be low for  
a minimum of 2 POCLK cycles. This may be turned off by setting  
the B1SELB inactive.  
PAROUTA  
PAROUTB  
PAROUTC  
PAROUTD  
LVTTL  
LVTTL  
O
O
158  
144  
96  
Parity Output. Odd or even parity depending on the input of parity  
select (PARSEL) for the 8 bit POUT[7:0] A, B, C, D and FP A, B, C,  
D (optional) data bus. PAROUT A, B, C, D, (FP A, B, C, D optional)  
is updated on the falling edge of POCLK.  
68  
LOS  
184  
Loss of Signal. Active high. When active, LOS indicates that a  
consecutive zero pattern for a minimum of 27usec of all zeros is  
detected on the incoming scrambled STS-48/STM-16 signal before  
descrambling. LOS is deactivated when two valid framing words are  
detected and no LOS is detected in between. LOS is updated on  
the falling edge of POCLK.  
Table 6. Common Pin Assignment and Descriptions  
Pin Name  
B1SELB  
Level  
I/O  
Pin #  
Description  
LVTTL  
I
203  
B1 Parity Byte Select. Active low. When active, B1 calculation and  
insertion in both the transmitter and receiver sections is enabled.  
When inactive, the B1 calculation and insertion in both the  
transmitter and receiver sections is disabled. This signal is static  
and must not be changed in normal operation.  
PARSEL  
RSTB  
LVTTL  
LVTTL  
I
I
197  
193  
Parity Select. When high selects even parity. When low selects  
odd parity. This signal is static and must not be changed in normal  
operation.  
Reset. Active low. Asyncronous reset input for the device. During  
reset PCLK A, B, C, D does not toggle and frame synchronization  
with SYNCRST will be activated.  
16  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Table 6. Common Pin Assignment and Descriptions (continued)  
Pin Name  
DLEB  
Level  
I/O  
Pin #  
Description  
LVTTL  
I
196  
Diagnostic Loopback Enable. Active low. When DLEB is inactive,  
the S3045 uses the primary data 311DATIN[7:0] and clock  
311CLKIN inputs. When active, the S3045 selects diagnostic  
loopback mode. In loopback mode the STS-48/STM-16  
transmitter outputs (311DATOUT[7:0] and 311CLKOUT) are  
internally connected to the STS-48/STM-16 receiver inputs  
(311DATIN[7:0] and 311CLKIN). This signal is static and must not  
be changed in normal operation.  
VDD  
202, 194, Power Pins. VDD pins must be tied to 3.3V. Note that it is  
159, 143, recommended that VDD be powered on after VDD5.  
123, 116,  
109, 97,  
82, 66,  
58, 56,  
55, 50,  
28, 22, 12  
VSS  
195, 188, Ground Pins. VSS pins must be tied to ground.  
170, 160,  
147, 142,  
132, 124,  
108, 98,  
90, 81,  
71, 67,  
63, 62,  
61, 60,  
59, 57,  
49, 40,  
29, 23,  
13, 3  
VDD5  
171, 131, TTL I/O Power Pins. There are three power pins that may be  
70  
connected to 3.3 volts or 5 volts. When these pins are connected  
to 3.3 volts, the TTL interface is a 3.3 volt LVTTL interface. When  
these pins are connected to 5 volts, the TTL inputs are 5 volt TTL  
tolerant. The TTL outputs are the same regardless of which  
voltage is tied to VDD5. All three pins must be connected to the  
same voltage level. Note that it is recommended that VDD5 be  
powered on before VDD.  
TEST_EN[0:1]  
TEST1  
LVTTL  
LVTTL  
LVTTL  
64, 65  
187  
TEST Enable. For factory test. For normal operation tie low.  
TEST Pins. For factory test. Must leave unconnected.  
TEST Pins. For factory test. Must leave unconnected.  
TEST2  
189  
17  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Figure 10. Parity Error (PARERR) Output Functional Timing Diagram  
Bad parity supplied to the S3045.  
PCLK A,B,C,D  
PARERR A,B,C,D  
Note: PARERR A,B,C,D output  
can be delayed by one to five  
PCLK A,B,C,D cycles.  
Figure 11. S3045 LVDS Inputs  
S3041/S3042  
S3045  
Z = 50  
o
P
N
P
100  
N
275Ω  
275Ω  
Figure 12. S3045 LVDS Outputs  
S3041  
S3045  
Z
= 50  
o
P
N
P
N
18  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Figure 13. OOF and PULSEOOF Functional Timing Diagram  
Frame  
Boundary  
/A1A2 A1A2 /A1A2 /A1A2 /A1A2 /A1A2 /A1A2 /A1A2 /A1A2 /A1A2  
A1A2  
A1A2  
OOF  
PULSEOOF  
Notes:  
– PULSEOOF minimum pulse width is 7.5ns.  
– Timing specifications are met with a 15pf load.  
19  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Figure 14. S3045 Pinout  
FRAMEP  
FRAMEN  
1
156  
155  
154  
153  
152  
151  
150  
149  
148  
147  
146  
145  
144  
143  
142  
141  
140  
139  
138  
137  
136  
135  
134  
133  
132  
131  
130  
129  
128  
127  
126  
125  
124  
123  
122  
121  
120  
119  
118  
117  
116  
115  
114  
113  
112  
111  
110  
109  
108  
107  
106  
105  
PINB[7]  
PINB[6]  
PINB[5]  
PINB[4]  
PINB[3]  
PINB[2]  
PINB[1]  
PINB[0]  
TIFPB  
2
3
VSS  
311DATINP[0]  
311DATINN[0]  
311DATINP[1]  
311DATINN[1]  
311DATINP[2]  
311DATINN[2]  
311DATINP[3]  
311DATINN[3]  
VDD  
4
5
6
7
8
9
10  
11  
12  
13  
14  
15  
16  
17  
18  
19  
20  
21  
22  
23  
24  
25  
26  
27  
28  
29  
30  
31  
32  
33  
34  
35  
36  
37  
38  
39  
40  
41  
42  
43  
44  
45  
46  
47  
48  
49  
50  
51  
52  
VSS  
PARINB  
PARERRB  
PAROUTB  
VDD  
VSS  
311DATINP[4]  
311DATINN[4]  
311DATINP[5]  
311DATINN[5]  
311DATINP[6]  
311DATINN[6]  
311DATINP[7]  
311DATINN[7]  
VDD  
VSS  
POCLKB  
POUTB[7]  
POUTB[6]  
POUTB[5]  
POUTB[4]  
POUTB[3]  
POUTB[2]  
POUTB[1]  
POUTB[0]  
VSS  
S3045 Pinout  
Top View  
208-pin PQFP/TEP  
28mm x 28mm x 3.4mm  
VSS  
311CLKINP  
311CLKINN  
311CLKOUTN  
311CLKOUTP  
VDD  
VDD5  
FPB  
PICLK  
VSS  
PCLKA  
PCLKB  
PCLKC  
PCLKD  
VSS  
311TCLKN  
311TCLKP  
311DATOUTN[0]  
311DATOUTP[0]  
311DATOUTN[1]  
311DATOUTP[1]  
311DATOUTN[2]  
311DATOUTP[2]  
311DATOUTN[3]  
311DATOUTP[3]  
VSS  
VDD  
SYNCRSTB  
PINC[7]  
PINC[6]  
PINC[5]  
PINC[4]  
PINC[3]  
VDD  
311DATOUTN[4]  
311DATOUTP[4]  
311DATOUTN[5]  
311DATOUTP[5]  
311DATOUTN[6]  
311DATOUTP[6]  
311DATOUTN[7]  
311DATOUTP[7]  
VSS  
PINC[2]  
PINC[1]  
PINC[0]  
TIFPC  
PARINC  
PARERRC  
VDD  
VSS  
POCLKC  
POUTC[7]  
POUTC[6]  
VDD  
PULSEN  
PULSEP  
20  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Figure 15. S3045 208 PQFP TEP (28 x 28 x 3.40mm) - Plastic Quad Flat Pack  
Top View  
Bottom View  
Note: The S3045 package is equipped with an embedded conductive heatsink on the top.  
Thermal Management  
Θja (Still Air)  
Θjc  
Device  
Max Package Power  
S3045  
2.20 W  
12.9 oC/W  
0.3 oC/W  
21  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Table 7. Absolute Maximum Ratings  
Parameter  
Min  
-65  
-0.5  
0
Typ  
Max Units  
Storage Temperature  
150  
+5.0  
+5.0  
˚ C  
V
Voltage on VDD/VDD5 with respect to GND  
Voltage on any LVPECL Input Pin  
V
VDD5  
+0.5  
Voltage on any LVTTL Input Pin  
-0.5  
V
ESD Ratings  
The S3045 is rated to the following ESD voltages based on the human body  
model:  
1. All pins are rated above 1000 V.  
Table 8. Recommended Operating Conditions  
Parameter  
Min  
0
Typ  
Max Units  
Ambient Temperature Under Bias  
Voltage on Vdd with respect to GND  
+70  
˚ C  
V
3.14  
3.3  
3.47  
VDD  
-2.00  
VDD  
-0.4  
Voltage on any LVPECL Input Pin  
Voltage on any LVTTL Input Pin  
V
V
VDD5  
+0.5V  
0
VDD  
+0.5V  
Voltage on any LVTTL Output Pin  
Power Dissipation  
V
1.66  
2.2  
W
Note: Power dissipation characterized using pseudo random pattern.  
22  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Table 9. LVPECL Input DC Characteristics  
Parameters  
VIL  
Description  
Input LOW Voltage  
Min  
VDD -2.0  
VDD -1.0  
-20  
Max  
VDD -1.6  
VDD -0.4  
20  
Units  
V
Conditions  
VIH  
IIH  
Input HIGH Voltage  
V
Single-ended input HIGH current  
Single-ended input LOW current  
SDVBB pin  
µA  
µA  
V
VIN = VDD -0.4V  
IIL  
-20  
20  
VIN = VDD -2.0V  
VOUT  
1.8  
2.2  
IOUT = 0, VDD = 3.47V  
Table 10. LVTTL Input/Output DC Characteristics  
Parameters Description  
VOH Output HIGH Voltage  
Min  
Typ  
Max Units  
Conditions  
2.4  
VDD  
.5  
V
V
VDD = 3.14V, IOH = -2.4mA  
VDD = 3.14V, IOL = 2.4mA  
VOL  
VIH  
Output LOW Voltage  
Input HIGH Voltage  
0.7  
VDD  
V
V
0.3  
VDD  
VIL  
Input LOW Voltage  
Input HIGH Current, for 5V tolerant  
input  
II  
1
50  
50  
10  
50  
-0.2  
mA  
µA  
µA  
µA  
µA  
V
V
IN = VDD5 = 5.5V  
VIN = 2.4V  
VIN = 2.4V  
IIH  
IIH  
IIL  
IIL  
VIK  
Input HIGH Current  
-50  
10  
2
Input HIGH Current with pulldowns  
TEST_EN0, TEST_EN1  
25  
Input LOW Current  
-500  
1
VIN = 0.5V  
VIN = 0.5V  
IIN = -18mA  
Input LOW Current with pulldowns  
TEST_EN0, TEST_EN1  
10  
Input Clamp Voltage  
-1.2  
-0.8  
Note: The term LVTTL implies CMOS levels for this device.  
23  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Table 11. LVDS Input/Output Characteristics  
Parameters  
IIH  
Description  
Input High Current  
Min  
-10  
Typ  
Max Units  
Conditions  
VIN = 1.4V  
10  
10  
µA  
µA  
mV  
V
IIL  
Input Low Current  
-10  
VIN = 1.0V  
VDIFF  
VIH  
VIL  
Ro  
Input Voltage Differential  
100  
0.95  
450  
1.75  
117  
V
Output Impedence Single ended  
57  
Ohm  
Ro mismatch between differential  
outputs  
Delta Ro  
1
%
VOS  
VOH  
VOL  
Output Offset Voltage  
Output High Voltage  
Output Low Voltage  
1.1  
1.2  
1.30  
1.45  
1.15  
1.40  
1.60  
1.25  
V
V
V
100 Ohms across diff. pair  
100 Ohms across diff. pair  
100 Ohms across diff. pair  
0.93  
VOUT  
(diff)  
Output Differential Voltage  
Input Differential Threshold  
250  
300  
450  
mV  
mV  
100 Ohms across diff. pair  
Vdim  
-100  
+100  
24  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Figure 16. AC Transmitter Parallel Input Data Timing Diagram  
t
PCLK  
PCLK A,B,C,D  
t
PICLK  
PICLK  
tS  
tH  
PIN  
PIN  
PINA[7:0]  
PINB[7:0]  
PINC[7:0]  
PIND[7:0]  
TIFP A,B,C,D  
PARIN A,B,C,D  
Table 12. AC Transmitter Parallel Data Input Timing Characteristics  
Parameters  
Characteristics  
PCLK A, B, C, D 77.76 MHz Output Clock  
PCLK A, B, C, D 77.76 MHz Output Clock duty cycle  
Parallel Input Clock (PICLK) period  
Min  
Typ  
Max Units  
tPCLK  
12.86  
ns  
40  
60  
60  
%
ns  
%
tPICLK  
12.86  
Parallel Input Clock (PICLK) duty cycle  
40  
PINA[7:0], PINB[7:0], PINC[7:0], PIND[7:0] PARIN A, B, C, D and  
TIFP A, B, C, D, data setup time with respect to rising edge of PICLK  
tSPIN  
tHPIN  
1.3  
ns  
PINA[7:0], PINB[7:0], PINC[7:0], PIND[7:0] PARIN A, B, C, D and  
TIFP A, B, C, D, data hold time with respect to rising edge of PICLK  
0.85  
ns  
ns  
PCLK Rise and Fall Times (10%–90%)  
2.5  
Notes  
1. Timing depicted in Figure 16 assumes PCLKA,B,C,D and PICLK are the same frequency.  
2. Timing specifications are met with a 15pf load.  
3. Timing specifications are measured with VDD = 3.3V at 25˚ C.  
25  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Figure 17. AC Parity Error Timing Diagram  
PCLK A,B,C,D  
tHERR  
tSERR  
PARERR A,B,C,D  
Table 13. Parity Error Timing Characteristics  
Symbol  
Description  
Min  
3.5  
3.5  
Max Units  
tSERR  
tHERR  
Parity Error Output Set-up Time w.r.t. PCLK  
Parity Error Output Hold Time w.r.t. PCLK  
ns  
ns  
Note: Timing specifications are met with a 15pf load.  
26  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Figure 18. AC Transmitter Output Timing Diagram  
311CLKOUTP  
tS  
311  
tH  
311  
311DATOUT[7:0]  
READP/N  
Notes:  
1. When a set-up time is specified on differential LVDS signals between an input and a clock, the set-up time is the time in nanoseconds from  
the cross-over point of the input to the cross-over point of the clock.  
2. When a hold time is specified on differential LVDS signals between an input and a clock, the hold time is the time in nanoseconds from the  
cross-over point of the clock to the cross-over point of the input.  
Table 14. Transmitter Output Timing Characteristics  
Symbol  
Description  
Min  
33  
Max Units  
311CLKOUT Duty Cycle  
67  
%
ns  
ns  
tS311  
tH311  
311DATOUT[7:0] Set-up Time w.r.t. 311CLKOUTP  
311DATOUT[7:0] Hold Time w.r.t. 311CLKOUTP  
0.65  
0.55  
Table 15. Transmitter Input Timing Characteristics  
Symbol Description  
311TCLK  
Min  
Max Units  
55  
311TCLK Duty Cycle  
45  
%
27  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Figure 19. AC Receiver Input Timing Diagram  
311CLKINP  
tS  
tH  
311O  
311O  
311DATIN[7:0]  
Frame  
Notes:  
1. When a set-up time is specified on differential LVDS signals between an input and a clock, the set-up time is the time in nanoseconds from  
the cross-over point of the input to the cross-over point of the clock.  
2. When a hold time is specified on differential LVDS signals between an input and a clock, the hold time is the time in nanoseconds from the  
cross-over point of the clock to the cross-over point of the input.  
Table 16. AC Receiver Input Timing Characteristics  
Symbol  
Description  
Min  
40  
Max Units  
311CLKIN Duty Cycle  
60  
%
ns  
ns  
tS311O  
tH311O  
311DATIN[7:0] and Frame Set-up Time w.r.t. 311CLKINP  
311DATIN[7:0] and Frame Hold Time w.r.t. 311CLKINP  
0.8  
0.8  
Figure 20. AC Pulse Timing Diagram  
311TCLKP  
PULSEP/N  
tS  
tH  
PS  
PS  
Table 17. AC Pulse Timing Characteristics  
Symbol  
Description  
Min  
Max Units  
tSPS  
tHPS  
PULSEP/N Setup Time w.r.t. 311TCLKP  
PULSEP/N Hold Time w.r.t. 311TCLKP  
0.8  
0.8  
ns  
ns  
28  
December 13, 1999 / Revision E  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
S3045  
Table 18. AC Receiver Output Timing Characteristics  
Symbol  
Description  
Min  
Max Units  
POCLK Frequency (nominally 77.76MHz)  
POCLK Duty Cycle  
78  
60  
MHz  
%
40  
4
tSPOUT  
tHPOUT  
POUT[7:0] Set-up Time to POCLK  
POUT[7:0] Hold Time to POCLK  
POCLK Rise and Fall Times (10% – 90%)  
ns  
3.5  
ns  
2.5  
ns  
Note: Timing specifications are met with a 15pf load. Timing specifications are measured with VDD = 3.3V at 25˚ C.  
Figure 21. AC Receiver Output Timing Diagram  
POCLKA, B, C, D  
tS  
tH  
POUT  
POUT  
POUT[7:0]A, B, C, D,  
FP, PAROUTA, B, C, D  
J0FP, B1ERR  
Table 19. Reset Timing Characteristics  
Symbol  
Description  
Min  
Max Units  
RSTB  
Minimum assertion of RESET(RSTB), to reset the S3045  
30  
ns  
29  
December 13, 1999 / Revision E  
S3045  
SONET/SDH OC-12 TO OC-48 MUX/DEMUX  
Ordering Information  
PREFIX  
DEVICE  
PACKAGE  
S – Integrated Circuit  
3045  
B – 208 PQFP/TEP  
X
XXXX  
X
Prefix Device  
Package  
Applied Micro Circuits Corporation • 6290 Sequence Dr., San Diego, CA 92121  
Phone: (858) 450-9333 • (800) 755-2622 • Fax: (858) 450-9885  
http://www.amcc.com  
AMCC reserves the right to make changes to its products or to discontinue any semiconductor product or service without notice, and  
advises its customers to obtain the latest version of relevant information to verify, before placing orders, that the information being relied  
on is current.  
AMCC does not assume any liability arising out of the application or use of any product or circuit described herein, neither does it  
convey any license under its patent rights nor the rights of others.  
AMCC reserves the right to ship devices of higher grade in place of those of lower grade.  
AMCC SEMICONDUCTOR PRODUCTS ARE NOT DESIGNED, INTENDED, AUTHORIZED, OR WARRANTED TO BE SUITABLE FOR  
USE IN LIFE-SUPPORT APPLICATIONS, DEVICES OR SYSTEMS OR OTHER CRITICAL APPLICATIONS.  
AMCC is a registered trademark of Applied Micro Circuits Corporation.  
Copyright ® 1999 Applied Micro Circuits Corporation  
30  
December 13, 1999 / Revision E  

相关型号:

S30460

Silicon Power Rectifier
MICROSEMI

S30460E3

Rectifier Diode, 1 Phase, 1 Element, 40A, 600V V(RRM), Silicon, DO-203AB, METAL, DO-5, 1 PIN
MICROSEMI

S30480

SILICON POWER RECTIFIER
MICROSEMI

S3049A

Interface Circuit, PQFP32, HEAT SINK, TQFP-32
ROCHESTER

S3049A

Interface Circuit, PQFP32, HEAT SINK, TQFP-32
AMCC

S3049B

Interface Circuit, DIE-42
ROCHESTER

S304GWA

Optoelectronic Device
EVERLIGHT

S304HWB

Optoelectronic Device
EVERLIGHT

S304RWB

Optoelectronic Device
EVERLIGHT

S304_07

Silicon Power Rectifier
MICROSEMI

S305

Medium/High Capacity Standard Size Toggles
NKK

S3050A

Clock Recovery Circuit, 1-Func, Bipolar, LLCC-32
AMCC