TI TLK6002ZEU

TLK6002
www.ti.com
SLLSE34 – MAY 2010
Dual Channel 0.47Gbps to 6.25Gbps Multi-Rate Transceiver
Check for Samples: TLK6002
1 Introduction
1.1
Transceiver Features
1
• Dual Channel 470Mbps to 6.25Gbps
Continuous/Multi-Rate Transceiver
• Supports all CPRI and OBSAI Data Rates
• Integrated Latency Measurement Function,
Accuracy of ±814 ps
• CPRI/OBSAI Automated Rate Sense (ARS)
Function
• Supports SERDES Operation, 8B/10B Data
Modes (20-bit and 16-bit + Controls)
• 20-bit HSTL Single-Ended Parallel Data
Interface (Integrated Source and End
Termination)
• Shared or Independent Reference Clock per
Channel
• Latency/Depth Configurable Transmit and
Receive FIFOs.
• Loopback Capability (Serial and Parallel Side),
OBSAI Compliant
• Supports Serial Retime Operation
• Supports PRBS (27–1), (223 – 1) and (231–1) and
CRPAT Long/Short Generation and Verification
1.2
•
•
•
•
•
• Dual Power Supply: 1.0V Core, and 1.5V/1.8V
I/O Nominal Supply
• Serial Side Three Tap Transmit De-emphasis
and Receive Adaptive Equalization to Allow
Extended Backplane Reach
• Programmable Output Swing on Serial Output
• Minimum Receiver Differential Input
Thresholds of 100mVdfpp
• Loss of Signal (LOS) detection (≤75 mVdfpp)
• Interface to Back Plane, Copper Cables, or
Optical Modules
• Hot Plug Protection
• JTAG; IEEE 1149.1 /1149.6 Test Interface
• MDIO; IEEE 802.3 Clause-22 Support
• 65nm Advanced CMOS Technology
• Industrial Ambient Operating Temp (–40°C to
85°C) at Full Rate
• Device Package; 324 PBGA
Applications
WI Infrastructure
CPRI and OBSAI Links
Proprietary Links
Backplane
High Speed Point- to-Point Transmission Systems
1.3
1.3.1
Overview
Device Description
The TLK6002 is a member of a portfolio of multi-gigabit transceivers, intended for use in ultra-high-speed
bi-directional point-to-point data transmission systems. It is specifically intended for base station RRH
(Remote Radio Head) application, but may also be used in other high speed applications. The TLK6002
supports a serial interface speed of 0.470 Gbps to 6.25 Gbps. Rate support includes all the CPRI and
OBSAI rates (0.6144/0.768/1.2288/1.536/2.4576/3.072/4.9152/6.144 Gbps) using a single fixed reference
clock frequency (either 122.88 MHz or 153.6 MHz).
1
Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of Texas
Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet.
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of the Texas
Instruments standard warranty. Production processing does not
necessarily include testing of all parameters.
Copyright © 2010, Texas Instruments Incorporated
TLK6002
SLLSE34 – MAY 2010
www.ti.com
This integrated circuit can be damaged by ESD. Texas Instruments recommends that all integrated circuits be handled with
appropriate precautions. Failure to observe proper handling and installation procedures can cause damage.
ESD damage can range from subtle performance degradation to complete device failure. Precision integrated circuits may be more
susceptible to damage because very small parametric changes could cause the device not to meet its published specifications.
TLK6002 20-bit parallel interface operates in 1.5V or 1.8V HSTL single-ended format. The 20-bit interface
allows low speed signals on the parallel side and therefore enabling the use of low cost FPGA in the
system design. The parallel interface can be programmed to be in SDR (Single Data Rate) or DDR
(Double Data Rate) modes. The line rate may be set to full (≤6.25Gbps), half (≤3.75Gbps), quarter
(≤1.88Gbps) or eighth (≤0.94Gbps). The line rate can be set using either device inputs or software control
registers.
The TLK6002 performs data conversion parallel-to-serial, serial-to-parallel and clock extraction as a
physical layer interface device. The serial transceiver interface operates at a maximum serial data rate of
6.25 Gbps.
TLK6002 accepts single-ended HSTL signals at its parallel transmit and receive data buses. If the internal
8B/10B coding and decoding are enabled, TDA/B_[19:0] are latched by TXCLK_A/B and sent to the
internal 8b/10b encoder, where the resulting encoded words are serialized and transmitted differentially
using a line clock derived from the SERDES reference clock at the desired line rate. If the internal coding
and decoding are disabled, TDA/B_[19:0] are defined as 20-bits of data being serialized and transmitted
unmodified according to the desired line rate.
The receive direction performs the serial-to-parallel conversion on the input serial data synchronizing the
resulting 20-bit parallel data to the recovered byte clock (RXCLK_A/B). The optionally decoded receive
data is available on the RDA/B_[19:0] output signals.
The serial transmitter and receiver are implemented using differential Current Mode Logic (CML) with
integrated termination resistors.
The TLK6002 provides two local (parallel side) and two remote (serial side) loopback modes for self-test
and system diagnostic purposes.
The TLK6002 has an integrated loss of signal (LOS) detection function, which is asserted in conditions
where the serial input signal does not have sufficient voltage amplitude (≤75 mVdfpp). Note that the input
signal must be ≥150 mVdfpp when loss of signal replacement of the receive datapath data is enabled
(register bit 6.6).
2
Introduction
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
8B/10B
Encoder
Parallel to
Serial
8B/10B
Encoder
T X C L K _A
20-bit
REGISTER
C h an n el A
T D A _[19:0 ]
Parallel to
Serial
TXAP
TXAN
T D B _[1 9:0]
T X C L K _B
20-bit
REGISTER
2x2
Mux
TXBP
TXBN
C h an n el B
R X C L K _A
20-bit
REGISTER
C h a n n el A
R D A _ [19:0]
COMMA
Detect &
8B/10B
Decoding
Serial to
Parallel
COMMA
Detect &
8B/10B
Decoding
Serial to
Parallel
RXAP
RXAN
R D B _[19:0]
R X C L K _B
20-bit
REGISTER
2x2
Mux
RXBP
RXBN
C h an n el B
Figure 1-1. TLK6002 Block Diagram
Introduction
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
3
TLK6002
SLLSE34 – MAY 2010
www.ti.com
TLK6002
Scrambler
TXCLK_A
20-bit register &
TX FIFO
TDA_[19:0]
10
10
8B/10B
Encoder
Pattern
Generator
CODEA_EN
Block Diagram Channel A
10
10-bit
With
pre-emphasis
TXAP
MUX
10-bit
TXAN
Parallel to
Serial
10-bit
Rate_A[2:0]
BIAS
Reference
CLock
PD_TRXA_N
Clock
Synthesizer
RESET_N
PRTAD[4:0]
MDC
MDIO
Bit Clk
MDIO
Interface
SCL
SDI
CS_N
SDO
TDI
TCK
TRST_N
TMS
Interpolator and
Clock Recovery
&
Adaptive EQ
Pattern
Verifier
JTAG
TDO
LOSA
RXCLK_A
20-bit
REGISTER
RDA_[19:0]
RX_FIFO
10-bit
PRBSA_PASS
10
COMMA
Detect &
8B/10B
Decoding
&
Descrambler
10
Serial
to
Parallel
EQ
RXAP
EQ
RXAN
Figure 1-2. TLK6002 Detail Block Diagram (Channel A)
4
Introduction
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
ARS
Control /
Divider
Incoming Serial Bit Rate *(1/2/4/8 Full/Half/Quarter/Eighth) / 4
Incoming Serial Bit Rate *(1/2/4/8 Full/Half/Quarter/Eighth) / 4
Note: Also equal in frequency to MPY*REFCLK/2
1
0
1/N
0.5:4
1
0
S/W
0.6
CLK_OUT_SEL
RXDA[9:0]
RXBCLK_A
S/W
0.1
REFCLK_A_SEL
RX
TXDA[9:0]
TX
REFCLKP/N
+
TXBCLK_A
Channel A
SERDES
REFCLK_0_P/N
0
1
-
+
+
CLK_OUT_P/N
RXAP
RXAN
TXAP
TXAN
ARS Channel A
ARS Channel B
+
0
1
-
RXDB[9:0]
RXBCLK_B
S/W
0.0
REFCLK_B_SEL
REFCLKP/N
TXDB[9:0]
TXBCLK_B
+
-
RX
TX
Channel B
SERDES
+
REFCLK_1_P/N
RXBP
RXBN
TXBP
TXBN
Legend:
= Primary Device Pin
S/W
x.x:x
= Software Programmable / Register Address.Bit
RXBCLK_*, TXBCLK_* frequency is Serial Bit Rate Divided by 10
RXBCLK_* is referred to as the recovered byte clock, and is always
synchronous with the incoming serial data rate (when valid).
Figure 1-3. TLK6002 Reference Clock/Output Clock Architecture
TXCLK_B
0
1
1
0
TXCLK_A
A S/W
3.10
TDA_[19:0]
TX A FIFO
TXDA_INT[9:0]
TXBCLK_A
CH A TX
TXDatapath
A Datapath
TXDA[9:0]
TXBCLK_A
Channel A
TX
Datapath
TXASERDES
TX B FIFO
TXDB_INT[9:0]
TXBCLK_B
CH B TX
TXDatapath
A Datapath
TXDB[9:0]
TXBCLK_B
Channel B
TX
Datapath
TXASERDES
0
1
1
0
TDB_[19:0]
B S/W
3.10
`
Legend:
= Primary Device Pin
S/W
x.x:x
= Software Programmable / Register Address.Bit
RXBCLK_*, TXBCLK_* frequency is Serial Bit Rate Divided by 10
Figure 1-4. TLK6002 Transmit Clock Architecture
Introduction
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
5
TLK6002
SLLSE34 – MAY 2010
www.ti.com
S/W
{3.9,3.8}
DDR:
SDR:
RXCLK_A
00
01
10
11
4
2
TXBCLK_A
RXBCLK_A
TXBCLK_B
RXBCLK_B
0
1
RDA_[19:0]
RX A FIFO
RXDA_INT[9:0]
RXBCLK_A
CH A RX
TXDatapath
A Datapath
RXDA[9:0]
RXBCLK_A
Channel A
TX
Datapath
RXASERDES
RX B FIFO
RXDB_INT[9:0]
RXBCLK_B
CH B RX
TXDatapath
A Datapath
RXDB[9:0]
RXBCLK_B
Channel B
RXASERDES
TX
Datapath
S/W
3.9
S/W
3.9
0
1
RDB_[19:0]
DDR:
SDR:
RXCLK_B
00
01
10
11
4
2
TXBCLK_B
RXBCLK_B
TXBCLK_A
RXBCLK_A
S/W
{3.9,3.8}
Legend:
= Primary Device Pin
S/W
x.x:x
= Software Programmable / Register Address.Bit
RXBCLK_*, TXBCLK_* frequency is Serial Bit Rate Divided by 10
Figure 1-5. TLK6002 Receive Clock Architecture
Line rates for OBSAI
Line rates for CPRI
614.4 /
1228. 8 /
2457. 6 /
3072 /
4915. 2 /
6144 Mbps
768 /
1536 /
3072 /
6144Mbps
ADS5232
ADS5240
ADS528 x
CPRI /
OBSAI
RF
I/Q
FPGA
DDC
GC 6016
12- bit
ADC
ADC
DR
TX _CLK
IF 2
61. 44 MHz
IF 1
61. 44 MHz
CDCEX52005
TLK 6002
( Serdes
LO _ 1
61.44MH z
TLK6002
TLK3131 /
32 /34
Serdes
X - tal
RX_CLK
245.76 MHz
61.44 MHz
FIR
I/Q
FPGA
Duplexer
LNA
16 - bit
DAC
DUC
GC 6016
0
FIR
Σ
90
PA
16 - bit
DAC
DAC5682 z
DAC 5688
DAC 5687
LO _ 1
Figure 1-6. TLK6002 Application Diagram
6
Introduction
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
2 Description
2.1
Pin Descriptions
Table 2-1. Pin Description – Signal Pins
Terminal
Signal
BGA
Direction
Type
Supply
Description
Channel A:
Serial Transmit Channel A Output. TXAP and TXAN comprise the transmit direction Channel A differential serial
high speed output signal.
TXAP
TXAN
V7
V6
Output
CML
AVDD
During device reset (RESET_N asserted low) these pins are driven differential zero.
These CML outputs must be AC coupled.
During pin based power down (PD_TRXA_N asserted low), these pins are floating.During register based power
down (1.15 asserted high), these pins are floating.
RXAP
RXAN
U9
U8
Input CML
AVDD
P1
Input HSTL
1.5V/1.8V
VDDQA
Serial Receive Channel A Input. RXAP and RXAN comprise the receive direction Channel A differential high
speed serial input signal. These signals must be AC coupled.
Transmit Input Channel A Clock. TXCLK_A is used to sample Channel A input parallel data (TDA_[19:0]).
TXCLK_A
This input must be synchronous (0 ppm) to the SERDES reference clock.
In SDR mode, this signal is equal in frequency to serial bit rate / 20.
In DDR mode, this signal is equal in frequency to serial bit rate / 40.
If unused in the application, this input must be grounded.
TDA_[19:0 ]
H1
J3
H2
K4
J2
J1
K5
L3
L5
M3
M4
K3
M5
N3
M2
P2
P4
N5
R2
P5
Parallel Input Channel A Transmit Data Bus.
These data signals are synchronous to and sampled by TXCLK_A.
Two data modes are supported, SDR (Single Data Rate), and DDR (Double Data Rate). SDR has two valid symbols
per TXCLK_A cycle, and DDR has four valid symbols per TXCLK_A cycle.
When input data is in 8b/10b encoded format (a.k.a. 20-bit data mode), TDA_[19:10] and TDA_[9:0] each carry a
symbol.
Input
HSTL
1.5V/1.8V
VDDQA
When input data is encoded internal to TLK6002 (8b/10b encoder enabled, a.k.a. 16-bit data mode), two symbols
are input at a time, defined as follows:
One Symbol – TDA_[8] contains the control bit (k-character indication) of data byte TDA_[7:0], and TDA_[9] is
unused and should be grounded.
Other Symbol – TDA_[18] contains the control bit (k-character indication) of data byte TDA_[17:10], and TDA_[19] is
unused and should be grounded.
Unused parallel input pins must be grounded.
See the following figures for more detail:
Figure 2-1 20-bit SDR Parallel Interface Mode.
Figure 2-2 16-bit SDR Parallel Interface Mode.
Figure 2-3 20-bit DDR Parallel Interface Mode
Parallel Channel A Receive Data Bus.
These output receive data signals are synchronous to RXCLK_A.
RDA_[19:0]
D1
B3
B1
C2
E2
F2
F3
A2
C3
D3
C1
F4
D4
E4
G4
G3
H4
E5
H5
G5
Two data modes are supported, SDR (Single Data Rate), and DDR (Double Data Rate). SDR has two valid symbols
per RXCLK_A cycle, and DDR has four valid symbols per RXCLK_A cycle.
When output data is in 8b/10b encoded format (a.k.a. 20-bit data mode), RDA_[19:10] and RDA_[9:0] each carry a
symbol.
When output data is decoded internal to TLK6002 (8b/10b decoder enabled, a.k.a. 16-bit data mode), two symbols
are output at a time, defined as follows:
Output
HSTL
1.5V/1.8V
VDDQA
One Symbol – RDA_[8] contains the control bit (k-character indication) of data byte RDA_[7:0], and RDA_[9]
indicates whether a 8b/10b disparity error was detected or an invalid code was received coincident with that
particular symbol
Other Symbol – RDA_[18] contains the control bit (k-character indication) of data byte RDA_[17:10], and RDA_[19]
indicates whether a 8b/10b disparity error was detected or an invalid code was received coincident with that
particular symbol.
During device reset (RESET_N asserted low) these pins are driven low.During pin based power down
(PD_TRXA_N asserted low), these pins are floating.During register based power down (1.15 asserted high), these
pins are floating.
See the following figures for more detail:
Figure 2-1 20-bit SDR Parallel Interface Mode.
Figure 2-2 16-bit SDR Parallel Interface Mode.
Figure 2-3 20-bit DDR Parallel Interface Mode.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
7
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 2-1. Pin Description – Signal Pins (continued)
Terminal
Signal
BGA
Direction
Type
Supply
Description
Channel A Rate select pins. These pins put channel A into one of the four supported (full/half/quarter/eighth)
channel operation rates, enable software control, or enable Auto Rate Sense (ARS):
000 – Full Rate mode
001 – Half Rate mode
010 – Quarter rate mode
011 – Eighth rate mode
100 – Software Selectable Rate (Recommended default board configuration)
101 – Channel A Auto Rate Sense (ARS) Function Enabled
Channel A SERDES settings are determined by Channel A ARS machine.
CLK_OUT_P/N selected by CLK_OUT_SEL
See Table 2-9 for additional details on CLK_OUT_P/N.
110 – Channel A Auto Rate Sense (ARS) Function Enabled
RATE_A[2:0]
V2
T4
U3
Input
LVCMOS
1.5V/1.8V
VDDO3
Channel A SERDES settings are determined by Channel A ARS machine.
CLK_OUT_P/N is not selected by CLK_OUT_SEL
Channel B may not be simultaneously configured with RATE_B=110
With respect to CLK_OUT_P/N, this setting has the highest priority.
See Table 2-9 for additional details on CLK_OUT_P/N.
111 – Channel A Auto Rate Sense (ARS) Function Enabled – Slave Mode
If Channel B ARS is enabled (RATE_B=101 or 110 only):
Channel A SERDES settings are determined by Channel B ARS machine.
CLK_OUT_P/N is not selected by CLK_OUT_SEL
See Table 2-9 for additional details on CLK_OUT_P/N.
If Channel B ARS is not enabled (RATE_B=000/001/010/011/111):
Channel A SERDES settings are determined by Channel A MDIO registers.
CLK_OUT_P/N selected by CLK_OUT_SEL
See Table 2-9 for additional details on CLK_OUT_P/N.
Channel A and B should not be in slave mode simultaneously. Both directions of Channel A are controlled by these
input signals.
The RATE_A[2] pin should be routed to an uninstalled header so that it could be driven externally in the event that
device debug is required. In application mode, it should be biased with a pull up or pull down resistor, and not
connected directly to a power or ground plane.
Receive Output Channel A Clock. RXCLK_A is synchronous to RDA_[19:0], and may be used externally to
sample Channel A output parallel data.
RXCLK_A
PRBSA_PASS
F1
V3
Output
HSTL
1.5V/1.8V
VDDQA
In SDR mode, this signal is equal in frequency to serial bit rate / 20.
In DDR mode, this signal is equal in frequency to serial bit rate / 40.
During device reset (RESET_N asserted low) this pin is driven low.
During pin based power down (PD_TRXA_N asserted low), this pin is floating.
During register based power down (1.15 asserted high), these pins are floating.
Receive PRBS Channel A Error Free (Pass) Indicator
When PRBS test is enabled (PRBS_EN=1):
PRBSA_PASS=1 indicates that PRBS pattern reception is error free.
Output
PRBSA_PASS=0 indicates that a PRBS error is detected.
LVCMOS
During device reset (RESET_N asserted low) this pin is driven low.
1.5V/1.8V
VDDO3 40Ω During pin based power down (PD_TRXA_N asserted low), this pin is floating.
During register based power down (1.15 asserted high), this pin is floating.
Driver
It is highly recommended that PRBSA_PASS be brought to easily accessible point on the application board
(header), in the event that debug is required.
CODEA_EN
LOSA
V4
U4
Input
LVCMOS
1.5V/1.8V
VDDO3
Encoder/Decoder Channel A Enable: When this pin is asserted high, the internal 8b/10b encoder/decoder is
enabled. This signal is OR’d with MDIO register bits, and should be pulled low through a resistor if software control
is desired. This pin should be routed to an uninstalled header so that it could be driven externally in the event that
device debug is required. In application mode, it should be biased with a pull up or pull down resistor, and not
connected directly to a power or ground plane.
Channel A Receive Loss Of Signal (LOS) Indicator.
LOSA = 0, signal detected.
LOSA = 1, Loss of signal (6.10 enabled).
Loss of signal detection is based on the input signal level.
When RXAP/N has an input signal of ≤75 mVdfpp, LOSA will be asserted (if enabled). The input signal should be ≥
150 mVdfpp for this function to operate reliably.
Output
LVCMOS
1.5V/1.8V
VDDO3 40Ω Other functions can be observed on LOSA realtime, configured via MDIO.
Driver
During device reset (RESET_N asserted low) this pin is driven low. During pin based power down (PD_TRXA_N
asserted low), this pin is floating. During register based power down (1.15 asserted high), this pin is floating.
It is highly recommended that LOSA be brought to easily accessible point on the application board (header), in the
event that debug is required.
8
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 2-1. Pin Description – Signal Pins (continued)
Terminal
Direction
Type
Supply
Signal
BGA
PD_TRXA_N
T6
Input
LVCMOS
1.5V/1.8V
VDDO3
U12
U13
Output
CML
AVDD
Description
Transceiver Power down. When this pin is held low (asserted), Channel A is placed in power down mode. When
deasserted , Channel A operates normally. After deassertion, a software datapath reset should be issued through
the MDIO interface.
Channel B:
TXBP
TXBN
RXBP
RXBN
V10
V11
Input
CML
AVDD
Serial Transmit Channel B Output. TXBP and TXBN comprise the transmit direction Channel B differential high
speed serial output signal. During device reset (RESET_N asserted low) these pins are driven differential zero.
These CML outputs must be AC coupled.
During pin based power down (PD_TRXB_N asserted low), these pins are floating. During register based power
down (1.15 asserted high), these pins are floating.
Serial Receive Channel B Input. RXBP and RXBN comprise the receive direction Channel B differential high
speed serial input signal. These signals must be AC coupled.
Channel B Rate select pins. These pins put channel B into one of the four supported (full/half/quarter/eighth)
channel operation rates, enable software control, or enable Auto Rate Sense (ARS):
000 – Full Rate mode
001 – Half Rate mode
010 – Quarter rate mode
011 – Eighth rate mode
100 – Software Selectable Rate (Recommended if ARS not used)
101 – Channel B Auto Rate Sense (ARS) Function Enabled
Channel B SERDES settings are determined by Channel B ARS machine.
CLK_OUT_P/N selected by CLK_OUT_SEL
See Table 2-9 for additional details on CLK_OUT_P/N.
110 – Channel B Auto Rate Sense (ARS) Function Enabled
RATE_B[2:0]
U16
U17
U18
Input
LVCMOS
1.5V/1.8V
VDDO2
Channel B SERDES settings are determined by Channel B ARS machine.
CLK_OUT_P/N is not selected by CLK_OUT_SEL
Channel A may not be simultaneously configured with RATE_A=110
With respect to CLK_OUT_P/N, this setting has the highest priority.
See Table 2-9 for additional details on CLK_OUT_P/N.
111 – Channel B Auto Rate Sense (ARS) Function Enabled – Slave Mode
If Channel A ARS is enabled (RATE_A=101 or 110 only):
Channel B SERDES settings are determined by Channel A ARS machine.
CLK_OUT_P/N is not selected by CLK_OUT_SEL
See Table 2-9 for additional details on CLK_OUT_P/N.
If Channel A ARS is not enabled (RATE_A=000/001/010/011/111):
Channel B SERDES settings are determined by Channel B MDIO registers.
CLK_OUT_P/N selected by CLK_OUT_SEL
See Table 2-9 for additional details on CLK_OUT_P/N.
Channel A and B should not be in slave mode simultaneously.
Both directions of Channel B are controlled by these input signals.
The RATE_B[2] pin should be routed to an uninstalled header so that it could be driven externally in the event that
device debug is required. In application mode, it should be biased with a pull up or pull down resistor, and not
connected directly to a power or ground plane.
Transmit Input Channel B Clock. TXCLK_B is used to sample Channel B input parallel data
(TDB_[19:0]).
TXCLK_B
T18
Input HSTL
1.5V/1.8V
VDDQB
This input must be synchronous (0 ppm) to the SERDES reference clock.
In SDR mode, this signal is equal in frequency to serial bit rate / 20.
In DDR mode, this signal is equal in frequency to serial bit rate / 40.
If unused in the application, this input must be grounded.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
9
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 2-1. Pin Description – Signal Pins (continued)
Terminal
Signal
BGA
TDB_[19:0]
J17
H17
J16
H18
K15
K16
L14
L16
M14
M16
M15
P17
N14
N16
J18
M17
R17
P15
P14
P18
Direction
Type
Supply
Description
Parallel Input Channel B Transmit Data Bus.
These data signals are synchronous to and sampled by TXCLK_B.
Two data modes are supported, SDR (Single Data Rate), and DDR (Double Data Rate). SDR has two valid symbols
per TXCLK_B cycle, and DDR has four valid symbols per TXCLK_B cycle.
When input data is in 8b/10b encoded format (a.k.a. 20-bit data mode), TDB_[19:10] and TDB_[9:0] each carry a
symbol.
Input HSTL
1.5V/1.8V
VDDQB
When input data is encoded internal to TLK6002 (8b/10b encoder enabled, a.k.a. 16-bit data mode), two symbols
are input at a time, defined as follows:
One Symbol – TDB_[8] contains the control bit (k-character indication) of data byte TDB_[7:0], and TDB_[9] is
unused and should be grounded.
Other Symbol – TDB_[18] contains the control bit (k-character indication) of data byte TDB_[17:10], and TDB_[19] is
unused and should be grounded.
Unused parallel input pins must be grounded.
See the following figures for more detail:
Figure 2-1 20-bit SDR Parallel Interface Mode
Figure 2-2 16-bit SDR Parallel Interface Mode
Figure 2-3 20-bit DDR Parallel Interface Mode
Parallel Channel B Receive Data Bus.These output receive data signals are synchronous to RXCLK_B.
RDB_[19:0]
D18
E17
B18
C17
F16
F17
A17
B16
C16
D16
C18
F15
A16
B15
D15
E15
G16
H15
G15
G14
Two data modes are supported, SDR (Single Data Rate), and DDR (Double Data Rate). SDR has two valid symbols
per RXCLK_B cycle, and DDR has four valid symbols per RXCLK_B cycle.
When output data is in 8b/10b encoded format (a.k.a. 20-bit data mode), RDB_[19:10] and RDB_[9:0] each carry a
symbol.
When output data is decoded internal to TLK6002 (8b/10b decoder enabled, a.k.a. 16-bit data mode), two symbols
are output at a time, defined as follows:
Output
HSTL
1.5V/1.8V
VDDQB
One Symbol - RDB_[8] contains the control bit (k-character indication) of data byte RDB_[7:0], and RDB_[9]
indicates whether a 8b/10b disparity error was detected or an invalid code was received coincident with that
particular symbol
Other Symbol - RDB_[18] contains the control bit (k-character indication) of data byte RDB_[17:10], and RDB_[19]
indicates whether a 8b/10b disparity error was detected or an invalid code was received coincident with that
particular symbol.
During device reset (RESET_N asserted low) these pins are driven low. During pin based power down
(PD_TRXB_N asserted low), these pins are floating. During register based power down (1.15 asserted high), these
pins are floating.
See the following figures for more detail:
Figure 2-1 20-bit SDR Parallel Interface Mode
Figure 2-2 16-bit SDR Parallel Interface Mode
Figure 2-3 20-bit DDR Parallel Interface Mode
Receive Output Channel B Clock. RXCLK_B is synchronous to RDB_[19:0], and may be used externally to
sample Channel B output parallel data.
RXCLK_B
F18
Output
HSTL
1.5V/1.8V
VDDQB
In SDR mode, this signal is equal in frequency to serial bit rate / 20.
In DDR mode, this signal is equal in frequency to serial bit rate / 40.
During device reset (RESET_N asserted low) this pin is driven low.
During pin based power down (PD_TRXB_N asserted low), this pin is floating.
During register based power down (1.15 asserted high), these pins are floating.
Receive PRBS Channel B Error Free (Pass) Indicator
When PRBS test is enabled (PRBS_EN=1):
PRBSB_PASS=1 indicates that PRBS pattern reception is error free.
PRBSB_PASS=0 indicates that a PRBS error is detected.
PRBSB_PASS
V18
Output
LVCMOS
1.5V/1.8V
During device reset (RESET_N asserted low) this pin is driven low.
VDDO2 40Ω During pin based power down (PD_TRXB_N asserted low), this pin is floating.
Driver
During register based power down (1.15 asserted high), this pin is floating.
It is highly recommended that PRBSB_PASS be brought to easily accessible point on the application board
(header), in the event that debug is required.
CODEB_EN
10
V16
Input
LVCMOS
1.5V/1.8V
VDDO2
Encoder/Decoder Channel B Enable: When this pin is asserted high, the internal 8b/10b encoder/decoder is
enabled. This signal is OR’d with MDIO register bits, and should be tied low if software control is desired.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 2-1. Pin Description – Signal Pins (continued)
Terminal
Signal
BGA
Direction
Type
Supply
Description
Channel B Receive Loss Of Signal (LOS) Indicator.
LOSB=0, signal detected.
LOSB=1, Loss of signal (6.10 enabled).
Loss of signal detection is based on the input signal level.
When RXBP/N has an input signal of ≤75 mVdfpp, LOSB will be asserted (if enabled). The input signal should be ≥
150 mVdfpp for this function to operate reliably.
LOSB
V17
Output
LVCMOS
1.5V/1.8V
Other functions can be observed on LOSB realtime, configured via MDIO.
VDDO2 40Ω
During device reset (RESET_N asserted low) this pin is driven low.
Driver
During pin based power down (PD_TRXB_N asserted low), this pin is floating.
During register based power down (1.15 asserted high), this pin is floating.
It is highly recommended that LOSB be brought to easily accessible point on the application board (header), in the
event that debug is required.
PD_TRXB_N
U15
Input
LVCMOS
1.5V/1.8V
VDDO2
Transceiver Power down. When this pin is held low (asserted), Channel B is placed in power down mode. When
deasserted , Channel B operates normally. After deassertion, a software datapath reset should be issued through
the MDIO interface.
Signals common to Channels A and B:
REFCLK_0_P/N
B5
A5
Input LVDS/
LVPECL
DVDD
Reference Clock Input Zero. This differential input is a clock signal used as a reference to either or both of the
bidirectional SERDES macros. It can be routed internally to either SERDES macro using device pins
(REFCLK_A_SEL and REFCLK_B_SEL) or through software registers. This input signal must be AC coupled. See
Figure 1-3. TLK6002 Reference Clock / Output Clock Architecture for more detail. If unused, REFCLK_0_P/N
should be pulled down to DGND through a shared 100Ω resistor.
REFCLK_1_P/N
C6
D6
Input LVDS/
LVPECL
DVDD
Reference Clock Input One. This differential input is a clock signal used as a reference to either or both of the
bidirectional SERDES macros. It can be routed internally to either SERDES macro using device pins
(REFCLK_A_SEL and REFCLK_B_SEL) or through software registers. This input signal must be AC coupled. See
Figure 1-3. TLK6002 Reference Clock / Output Clock Architecture for more detail. If unused, REFCLK_1_P/N
should be pulled down to DGND through a shared 100Ω resistor
REFCLK_A_SEL
R6
Input
LVCMOS
1.5V/1.8V
VDDO3
Reference Clock Select Channel A. This input, when low, selects REFCLK_0_P/N as the clock reference to
Channel A SERDES macro. When high, REFCLK_1_P/N is selected as the clock reference to Channel A SERDES
macro. If software control is desired (register bit 0.1), this input signal should be tied low. See Figure 1-3. TLK6002
Reference Clock / Output Clock Architecture for more detail.
REFCLK_B_SEL
T2
Input
LVCMOS
1.5V/1.8V
VDDO3
Reference Clock Select Channel B. This input, when low, selects REFCLK_0_P/N as the clock reference to
Channel B SERDES macro. When high, REFCLK_1_P/N is selected as the clock reference to Channel B SERDES
macro. If software control is desired (register bit 0.0), this input signal should be tied low. See Figure 1-3. TLK6002
Reference Clock/Output Clock Architecture for more detail.
PRBS_EN
CLK_OUT_P/N
R16
J6
J7
Input
LVCMOS
1.5V/1.8V
VDDO2
Output CML
DVDD
Enable PRBS: When this pin is asserted high, the internal PRBS generator and verifier circuits are enabled on both
transmit and receive data paths of both channels.
This signal is logically OR’d with an mdio register bit.
PRBS 231-1 is selected by default, and can be changed in MDIO register 7.10:8.
Note that PRBS is not possible in eighth rate mode.
The PRBS_EN pin should be routed to an uninstalled header so that it could be driven externally in the event that
device debug is required. In application mode, it should be biased with a pull up or pull down resistor (or allow for
an isolation mechanism from the on board driver), and not connected directly to a power or ground plane.
Recovered Byte Clock. If ARS is not enabled, and CLK_OUT_SEL is low, an optionally divided version of Channel
A recovered byte clock is output onto CLK_OUT_P/N. If ARS is not enabled, and CLK_OUT_SEL is high, an
optionally divided version of Channel B recovered byte clock is output onto CLK_OUT_P/N. The recovered byte
clock is synchronous to the incoming serial data rate for the selected channel. See Figure 1-3. TLK6002 Reference
Clock/Output Clock Architecture for more detail. The recovered byte clock can be divided by one, two, four, or eight
as selected in an mdio register.
If ARS is enabled, the CLK_OUT_P/N output is selected via Table 2-9.
This CML output must be AC coupled.
During device reset (RESET_N asserted low) this pin is driven differential zero. During pin based power down
(PD_TRXA_N and PD_TRXB_N asserted low), these pins are floating.
During register based power down (1.15 asserted high both channels), these pins are floating.
CLK_OUT_SEL
T15
Input
LVCMOS
1.5V/1.8V
VDDO2
Output Clock Selection. If ARS is not enabled and CLK_OUT_SEL is low, Channel A recovered byte clock is
output onto CLK_OUT_P/N. If ARS is not enabled and CLK_OUT_SEL is high, Channel B recovered byte clock is
output onto CLK_OUT_P/N. If software control is desired (register bit 0.6), this input signal should be tied low. See
Figure 1-3. TLK6002 Reference Clock / Output Clock Architecture for more detail. If ARS is enabled, the function of
CLK_OUT_SEL is shown in Table 2-9.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
11
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 2-1. Pin Description – Signal Pins (continued)
Terminal
Signal
BGA
Direction
Type
Supply
Description
Port Address. Used to select the Port ID.
PRTAD[4:1] selects the device port address. TLK6002 has two different PHY addresses (ports). Selecting a unique
PRTAD[4:1] per TLK6002 device allows 16 TLK6002 devices per MDIO bus. Each channel can be accessed by
setting the appropriate port address field within the serial interface protocol transaction.
PRTAD[4:0]
V15
M8
K12
K11
L11
Input
LVCMOS
1.5V/1.8V
VDDO2/
VDDO1/
VDDO1/
VDDO1/
VDDO1
TLK6002 will respond if the 4 MSB’s of the inband PHY address field on MDIO protocol (PA[4:1]) matches
PRTAD[4:1]. The LSB of PHY address field (PA[0]) will determine which channel/port within TLK6002 to respond to.
PRTAD[0] is not used functionally, but is present for device testability and compatibility with other devices in the
family of products.
Channel A responds to port address 0 within the block of two port addresses.
Channel B responds to port address 1 within the block of two port addresses.
PRTAD[0] should be grounded on the application board.
The PRTAD[3] pin in application mode should be biased with a pull up or pull down resistor (or allow for an isolation
mechanism from the on board driver), and not connected directly to a power or ground plane. The application board
should allow the flexibility of easily reworking the PRTAD[3] signal to a high level if device debug is necessary (by
including an uninstalled resistor to VDDO1).
RESET_N
MDC
MDIO
SCL
SDO
SDI
CS_N
12
V1
Input
LVCMOS
1.5V/1.8V
VDDO3
Low True Device Reset. When asserted (low logic level), this signal resets the entire TLK6002 device. RESET_N
must be held asserted for at least 10 µS after device power stabilization.
MDIO clock input. Clock input for the Clause 22 MDIO interface.
Note that an external pullup is generally not required on MDC.
T1
Input
LVCMOS
w/Hysteresi
s 1.5V/1.8V
VDDO3
U2
Input/
Output
LVCMOS
1.5V/1.8V
VDDO3 25Ω
Driver
MDIO data I/O. MDIO interface data input/output signal for the Clause 22 MDIO interface.
This signal must be externally pulled up to VDDO3, using a 2 kΩ resistor.
During device reset (RESET_N asserted low) this pin is floating. During software initiated power down the
management interface remains active for control register writes and reads. Certain status bits are not deterministic
as their generating clock source may be disabled as a result of asserting either power down input signal. During pin
based power down (PD_TRXA_N and PD_TRXB_N asserted low), this pin is floating. During register based power
down (1.15 asserted high both channels), this pin is driven normally.
H13
Input/
Output
LVCMOS
1.5V/1.8V
VDDO1 25Ω
Driver
SPI Clock (SPI_CLK). Defaults to Output, Driven Low. Can be used as a SPI interface or a generic customer
controllable I/O interface. When used as part of the SPI interface, this signal is the SPI clock to be used with
external TI Jitter cleaner or clock Distribution device.
Three register bits (15.14:12) control this I/O signal. See the detailed register bit description for operational detail.
If unused in the application, this signal can be left floating.
Careful programming is required to prevent accidental contention with simultaneous external drivers. During device
reset (RESET_N asserted low) this pin is driven low. During pin based power down (PD_TRXA_N and PD_TRXB_N
asserted low), this pin is floating. During register based power down (1.15 asserted high both channels), these pins
are driven per register setting
K13
Input/
Output
LVCMOS
1.5V/1.8V
VDDO1 25Ω
Driver
SPI Data. Defaults to Input. Can be used as a SPI interface or a generic customer controllable I/O interface. When
used as part of the SPI interface, this signal is the SPI data from the external TI Jitter cleaner or clock Distribution
device to the TLK6002.
Three register bits (15.10:8) control this I/O signal. See the detailed register bit description for operational detail.
If unused in the application, this signal should be pulled to ground. Careful programming is required to prevent
accidental contention with simultaneous external drivers.
During device reset (RESET_N asserted low) this pin is floating. During pin based power down (PD_TRXA_N and
PD_TRXB_N asserted low), this pin is floating. During register based power down (1.15 asserted high both
channels), these pins are driven per register setting.
E14
Input/
Output
LVCMOS
1.5V/1.8V
VDDO1 25Ω
Driver
SPI Data. Defaults to Output, driven low. Can be used as a SPI interface or a generic customer controllable I/O
interface. When used as part of the SPI interface, this signal is the SPI data from TLK6002 to the external TI Jitter
cleaner or clock Distribution device.
Three register bits (15.6:4) control this I/O signal. See the detailed register bit description for operational detail.
If unused in the application, this signal can be left floating.
Careful programming is required to prevent accidental contention with simultaneous external drivers. During device
reset (RESET_N asserted low) this pin is driven low.
During pin based power down (PD_TRXA_N and PD_TRXB_N asserted low), this pin is floating. During register
based power down (1.15 asserted high both channels), these pins are driven per register setting.
D14
Input/
Output
LVCMOS
1.5V/1.8V
VDDO1 25Ω
Driver
SPI Chip Select. Defaults to Output, Driven High. Can be used as a SPI interface or a generic customer
controllable I/O interface. When used as part of the SPI interface, this signal is the chip select for the external TI
Jitter cleaner or clock Distribution device. Low=Select Device. High=Device Not Selected.
Three register bits (15.2:0) control this I/O signal. See the detailed register bit description for operational detail.
If unused in the application, this signal can be left floating. Careful programming is required to prevent accidental
contention with simultaneous external drivers.
During device reset (RESET_N asserted low) this pin is driven high. During pin based power down (PD_TRXA_N
and PD_TRXB_N asserted low), this pin is floating. During register based power down (1.15 asserted high both
channels), these pins are driven per register setting.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 2-1. Pin Description – Signal Pins (continued)
Terminal
Signal
BGA
Direction
Type
Supply
Description
R15
Input
LVCMOS
1.5V/1.8V
VDDO2
(Internal
Pullup)
JTAG Input Data. TDI is used to serially shift test data and test instructions into the device during the operation of
the test port. In system applications where JTAG is not implemented, this input signal may be left floating.
During pin based power down (PD_TRXA_N and PD_TRXB_N asserted low), this pin is not pulled up. During
register based power down (1.15 asserted high both channels), this pin is pulled up normally.
JTAG Output Data. TDO is used to serially shift test data and test instructions out of the device during operation of
the test port. When the JTAG port is not in use, TDO is in a high impedance state.
R14
Output
LVCMOS
1.5V/1.8V
VDDO2 50Ω
Driver
R3
Input
LVCMOS
1.5V/1.8V
VDDO3
(Internal
Pullup)
JTAG Mode Select. TMS is used to control the state of the internal test-port controller. In system applications
where JTAG is not implemented, this input signal can be left unconnected.
During pin based power down (PD_TRXA_N and PD_TRXB_N asserted low), this pin is not pulled up. During
register based power down (1.15 asserted high both channels), this pin is pulled up normally.
JTAG Clock. TCK is used to clock state information and test data into and out of the device during boundary scan
operation. In system applications where JTAG is not implemented, this input signal should be grounded.
T16
Input
LVCMOS
w/Hysteresi
s 1.5V/1.8V
VDDO2
TRST_N
T3
Input
LVCMOS
1.5V/1.8V
VDDO3
(Internal
Pulldown)
JTAG Test Reset. TRST_N is used to reset the JTAG logic into system operational mode. This input can be left
unconnected in the application and is pulled down internally, disabling the JTAG circuitry. If JTAG is implemented
on the application board, this signal should be deasserted (high) during JTAG system testing, and otherwise
asserted (low) during normal operation mode.
During pin based power down (PD_TRXA_N and PD_TRXB_N asserted low), this pin is not pulled down. During
register based power down (1.15 asserted high both channels), this pin is pulled down normally.
TESTEN
T17
Input
LVCMOS
1.5V/1.8V
VDDO2
Test Enable. This signal is used during the device manufacturing process. It should be grounded through a resistor
in the device application board. The application board should allow the flexibility of easily reworking this signal to a
high level if device debug is necessary (by including an uninstalled resistor to VDDO2).
GPI0
R4
Input
LVCMOS
1.5V/1.8V
VDDO3
General Purpose Input Zero. This signal is used during the device manufacturing process. It should be grounded
through a resistor on the device application board. The application board should also allow the flexibility of easily
reworking this signal to a high level if device debug is necessary (by including an uninstalled resistor to VDDO3).
GPI1
K10
Input
LVCMOS
1.5V/1.8V
VDDO1
General Purpose Input One. This signal can be used to logically combine an external status condition with LOSA
or LOSB if enabled in an mdio register. Note that if GPI1 is low, LOSA/B will be asserted if logical combination is
enabled. Similarly, if GPI1 is high, LOSA/B will be deasserted. If unused, this input should be grounded in the
device application (not floating).
AMUXA
U5
Analog I/O
SERDES Channel A Analog Testability I/O. This signal is used during the device manufacturing process. It should
be left unconnected in the device application.
AMUXB
V14
Analog I/O
SERDES Channel B Analog Testability I/O. This signal is used during the device manufacturing process. It should
be left unconnected in the device application.
RESRA,
RESTA,
RESRB,
RESTB
K8
M1
J14
M18
Analog
Input
HSTL Impedance Matching Resistors. These resistors are used as a reference for internal terminations on the
HSTL inputs and outputs. Each RES* pin requires it’s own resistor, sharing resistors between RES* pins is not
possible. A 50 ohm 0.5% tolerance resistor should be selected to guarantee device datasheet specified parallel
interface timing specification.
TDI
TDO
TMS
TCK
During device reset (RESET_N asserted low) this pin is floating. During pin based power down (PD_TRXA_N and
PD_TRXB_N asserted low), this pin is floating. During register based power down (1.15 asserted high both
channels), this pin is floating.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
13
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 2-2. Pin Description – Power Pins
Terminal
Type
Description
Signal
BGA
VDDD
T8
T11
Power
SERDES Digital logic power
Provides power for digital circuitry internal to the SERDES. 1.0V nominal.
AVDD
U6
T9
T10
U11
T13
U14
Power
SERDES Analog Power
AVDD provides supply voltage for the high-speed analog circuits. 1.0V nominal.
DVDD
L6
L9
L13
M10
M12
N6
N13
R8
R10
R12
T5
Power
Digital Core Power
DVDD provides supply voltage to the digital core. 1.0V nominal.
VDDT
V8
V12
Power
SERDES Analog Power
VDDT provides supply voltage for the high-speed analog circuits, termination voltage. 1.0V nominal.
VDDRA/B
T7
T12
Power
SERDES Analog Regulator Power
VDDRA and VDDRB provide supply voltage for the internal PLL regulator. 1.5V or 1.8V nominal.
VDDQA/B
A3
B2
B17
C4
C15
E1
E18
F5
F14
G2
G17
J4
J15
K1
K18
L4
L7
L8
L12
L15
L17
M6
M13
N1
N18
P3
P6
P13
P16
Power
HSTL I/O Power
VDDQA and VDDQB provide supply voltage for the HSTL inputs and outputs. 1.5V or 1.8V nominal.
VDDO1/2/3
L10
R13
R7
Power
LVCMOS I/O Power
VDDO1, VDDO2, and VDDO3 provide supply voltage for the LVCMOS inputs and outputs. 1.5V or 1.8V nominal.
VPP
K9
Power
Factory Program Voltage
Programming supply voltage for TI internal use during device manufacturing. The application must connect this power
supply directly to DVDD.
AGND
R9
R11
T14
U7
U10
V5
V9
V13
Ground
Analog Ground
Analog ground.
14
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 2-2. Pin Description – Power Pins (continued)
Terminal
Type
Description
Signal
BGA
DGND
A1
A4
A6
A15
A18
B4
B6
C5
D2
D5
D17
E3
E16
G1
G18
H3
H14
H16
J5
K2
K6
K14
K17
L2
M7
M9
M11
N2
N4
N7
N8
N10
N11
N12
N15
N17
P7
P8
P9
P10
P11
P12
R1
R5
R18
U1
Ground
Digital Ground
Digital ground
VREFTA,
VREFTB
L1
L18
Voltage
Reference
HSTL Voltage Reference
These high impedance voltage reference inputs are used as a signal comparison level for HSTL input signals. These
signals should be created using a resistive voltage divider (dual 1kΩ) between VDDQA or VDDQB and DGND. These
signals should be locally decoupled as close to the device pins as possible.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
15
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 2-2. Pin Description – Power Pins (continued)
Terminal
Type
Description
Signal
BGA
NC
A7
A8
A9
A10
A11
A12
A13
A14
B7
B8
B9
B10
B11
B12
B13
B14
C7
C8
C9
C10
C11
C12
C13
C14
D7
D8
D9
D10
D11
D12
D13
E6
E7
E8
E9
E10
E11
E12
E13
F6
F7
F8
F9
F10
F11
F12
F13
G6
G7
G8
G9
G10
G11
G12
G13
H6
H7
H8
H9
H10
H11
H12
J8
J9
J10
J11
J12
J13
K7
No
Connect
No Connect
These BGAs can be left unconnected in the application
NC33
N9
Reserved
Input
Reserved
This input pin should be connected to DVDD through a zero ohm resistor in the device application.
16
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
2.2
SLLSE34 – MAY 2010
Device Pinout Diagram
Table 2-3. Device Pinout Diagram – (Top View)
1
2.3
2
3
4
DGND
5
6
REFCLK_0
_N
REFCLK_0
_P
A
DGND
RDA_12
VDDQA
B
RDA_17
VDDQA
RDA_18
DGND
C
RDA_9
RDA_16
RDA_11
VDDQA
DGND
D
RDA_19
DGND
RDA_10
RDA_7
DGND
E
VDDQA
RDA_15
DGND
RDA_6
RDA_2
F
RXCLK_A
RDA_14
RDA_13
RDA_8
VDDQA
G
DGND
VDDQA
RDA_4
RDA_5
RDA_0
H
TDA_19
TDA_17
DGND
RDA_3
RDA_1
J
TDA_14
TDA_15
TDA_18
VDDQA
DGND
K
VDDQA
DGND
TDA_8
TDA_16
TDA_13
7
DGND
CGND
DGND
VDD_XTAL
REFCLK_1
_P
REFCLK_1
_N
VDD_CP1_
PFD1
VDD_PRI_I
N1
VDD_FBIN
8
XTALN
LF1B
VDD_SREF
LF1A
CGND
VDD_PLL1
9
XTALP
CGND
VDD_PRI_I
N2
VDD_SEC_
IN2
SYNTHREF
10
PRI_REF2P
PRI_REF2N
11
12
13
14
CGND
CGND
RC1
RC3
15
DGND
16
RDB_7
17
18
RDB_13
DGND
RDB_17
VDD_PLL2
LF2B
AMUX
RC2
RDB_6
RDB_12
VDDQB
CGND
VDD_CP2_
PFD2
LF2A
CGND
CMSEL
VDDQB
RDB_11
RDB_16
RDB_9
CGND
CGND
CGND
RFOUTN
CS_N
RDB_5
RDB_10
DGND
RDB_19
CGND
CGND
CGND
CGND
VDD_VCO
RFOUTP
SDI
RDB_4
DGND
RDB_18
VDDQB
FBINN
CGND
CGND
CGND
CGND
VDD_RF_P
S
VDD18
VDDQB
RDB_8
RDB_15
RDB_14
RXCLK_B
VDD_DIG_
IO
FBINP
CGND
CGND
CGND
CGND
CGND
VDD_SEC_
IN1
CLK_OUT_
P
CGND
Y3N
Y3P
Y0N
Y0P
CGND
RDB_0
RDB_1
RDB_3
VDDQB
DGND
SCL
DGND
RDB_2
DGND
TDB_18
TDB_16
CLK_OUT_
N
VDD_Y4
VDD_Y3
VDD_Y2
VDD_Y1
DGND
RESETN
RESRA
VPP
GPI1
PRTAD1
VDD_Y0
PDN
RESRB
VDDQB
TDB_17
TDB_19
TDB_5
PRTAD2
SDO
DGND
TDB_15
TDB_14
DGND
VDDQB
VREFTB
L
VREFTA
DGND
TDA_12
VDDQA
TDA_11
DVDD
VDDQA
VDDQA
DVDD
VDDO1
PRTAD0
VDDQB
DVDD
TDB_13
VDDQB
TDB_12
VDDQB
M
RESTA
TDA_5
TDA_10
TDA_9
TDA_7
VDDQA
DGND
PRTAD3
DGND
DVDD
DGND
DVDD
VDDQB
TDB_11
TDB_9
TDB_10
TDB_4
RESTB
N
VDDQA
DGND
TDA_6
DGND
TDA_2
DVDD
DGND
DGND
VDD33
DGND
DGND
DGND
DVDD
TDB_7
DGND
TDB_6
DGND
VDDQB
P
TXCLK_A
TDA_4
VDDQA
TDA_3
TDA_0
TDB_0
R
DGND
TDA_1
TMS
GPI0
DGND
T
MDC
REFCLK_B
_SEL
TRST_N
RATE_A1
DVDD
VDDQA
REFCLK_A
_SEL
PD_TRXA_
N
DGND
DGND
DGND
DGND
DGND
DGND
VDDQB
TDB_1
TDB_2
VDDQB
TDB_8
VDDO3
DVDD
AGND
DVDD
AGND
DVDD
VDDO2
TDO
TDI
PRBS_EN
TDB_3
DGND
VDDRA
VDDD
AVDD
AVDD
VDDD
VDDRB
AVDD
AGND
TCK
TESTEN
TXCLK_B
RATE_B2
RATE_B1
RATE_B0
CODEB_EN
LOSB
PRBSB_PA
SS
U
DGND
MDIO
RATE_A0
LOSA
AMUXA
AVDD
AGND
RXAN
RXAP
AGND
AVDD
TXBP
TXBN
AVDD
V
RESET_N
RATE_A2
PRBSA_PA
SS
CODEA_EN
AGND
TXAN
TXAP
VDDT
AGND
RXBP
RXBN
VDDT
AGND
AMUXB
CLK_OUT_
SEL
PD_TRXB_
N
PRTAD4
CPRI/OBSAI Specific Operation Modes
The TLK6002 contains an internal low-jitter high quality oscillator that is used as a frequency multiplier for
the serdes and other internal circuits of the device. The rate pins (and mdio registers) as well as the
SERDES PLL multiplier are used to program the line rate and the REFCLK frequency for various
applications. See Appendix B for more details on SERDES reference clock, rate, and multiplier selection
(rates beyond the CPRI/OBSAI specific rates).
The TLK6002 is optimized for operation at a serial data rate of 470 Mbit/s through 6.25 Gbit/s. The
external differential reference clock has a large operating frequency range allowing support for many
different applications. The reference clock frequency must be within ±200 PPM of the incoming serial data
rate (±100 PPM of nominal data rate), and have less than 40ps of jitter. Table 2-4 and Table 2-5 show a
summary of frequency ranges used for the CPRI and OBSAI applications. The transmit parallel input clock
must be frequency locked (0 ppm) to the supplied/selected reference clock (REFCLK_0/1_P/N) frequency.
Table 2-4. CPRI Line Rate Selection (1)
(1)
LINE RATE
(Mbps)
SERDES PLL
MULTIPLIER VALUE
RATE SELECT (PINS OR
REGISTER VALUE)
TXCLK_A/B
(MHz)
REFCLKP/N
(MHz)
6144.00
20/25
4915.20
16/20
Full
307.2
153.60/122.88
Full
245.76
153.60/122.88
3072.00
20/25
2457.60
16/20
Half
153.6
153.60/122.88
Half
122.88
1228.80
153.60/122.88
16/20
Quarter
61.44
153.60/122.88
614.40
16/20
Eighth
30.72
153.60/122.88
In DDR mode TX_CLK frequencies will be half the values in the table above. The table above indicate two possible REFCLK
frequencies, 153.60MHz and 122.88MHz which can be used based on the application preference. The Serdes PLL Multiplier (MPY) has
been given for each REFCLK frequency respectively. Note that Channel A and B are independent, and their application rates and
references clocks are separate.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
17
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 2-5. OBSAI Line Rate Selection (1)
(1)
LINE RATE
(Mbps)
SERDES PLL
MULTIPLIER VALUE
RATE SELECT (PINS OR
REGISTER VALUE)
TXCLK_A/B
(MHz)
REFCLKP/N
(MHz)
6144.00
20/25
Full
307.2
153.60/122.88
3072.00
20/25
Half
153.6
153.60/122.88
1536.00
20/25
Quarter
76.8
153.60/122.88
768.00
20/25
Eighth
38.4
153.60/122.88
In DDR mode TX_CLK frequencies will be half the values in the table above. The table above indicate two possible REFCLK
frequencies, 153.60MHz and 122.88MHz which can be used based on the application preference. The Serdes PLL Multiplier (MPY) has
been given for each REFCLK frequency respectively. Note that Channel A and B are independent, and their application rates and
references clocks are separate.
2.4
2.4.1
Parallel Interface Modes
20-bit SDR (Single Data Rate) Mode (8b/10b Encoder/Decoder Disabled)
Channel A TX: TDA_[19:0] → TXAP/N (using TXCLK_A). RX: RXAP/N → RDA_[19:0] (using RXCLK_A).
Channel B TX: TDB_[19:0] → TXBP/N (using TXCLK_B). RX: RXBP/N → RDB_[19:0] (using RXCLK_B).
20 Bits (two symbols) of already encoded (TX) or undecoded (RX) data are transferred per parallel
interface clock cycle. Symbols are defined by a group of 10 parallel bits. Note that four symbols are shown
in Figure 2-1: Data0[19:10], Data0[9:0], Data1[19:10], Data1[9:0].
Symbol Transmission Order:
When 3.5/3.4 = 0: Data0[19:10] is the first transmitted or received symbol. Data0[9:0] is next, then
Data1[19:10].
When 3.5/3.4 = 1: Data0[9:0] is the first transmitted or received symbol. Data0[19:0] is next, then
Data1[9:0].
Bit Transmission Order within a Symbol:
When 8.3/2 = 0, Data[19] or Data[9] bits are serially transmitted first or received first respectively per
symbol.
When 8.3/2 = 1, Data[10] or Data[0] bits are serially transmitted first or received first respectively per
symbol.
18
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
SDR Rising Edge Aligned Timing
TXCLK_A/B
TDA/B_[19:0]
Data0[19:0]
Data1[19:0]
RXCLK _A /B
RDA/B_ [19:0]
Data0[19:0]
Data1[19:0]
SDR Falling Edge Aligned Timing
TXCLK_A/B
Data0[ 19:0]
TDA/B_[19:0]
Data1[ 19:0]
RXCLK _A /B
Data0[19:0]
RDA /B_ [ 19:0]
Data1[19:0]
Figure 2-1. 20-bit SDR Parallel Interface Mode
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
19
TLK6002
SLLSE34 – MAY 2010
2.5
www.ti.com
16-bit SDR (Single Data Rate) Mode (8b/10b Encoder/Decoder Enabled)
Channel A TX: TDA_[18:10,8:0] → TXAP/N (Using TXCLK_A). RX: RXAP/N → RDA_[19:0] (Using
RXCLK_A).
Channel B TX: TDB_[18:10,8:0] → TXBP/N (Using TXCLK_B). RX: RXBP/N → RDB_[19:0] (Using
RXCLK_B).
The 16 Bits (two symbols) of unencoded (TX) or decoded (RX) data are transferred per parallel interface
clock cycle. Symbols are defined by a group of 9 parallel bits comprising of a K character control bit and a
byte of data (plus a high true disparity error or invalid symbol bit in RX only on RD*_[19] and RD*_[9]).
Please note that four symbols are shown in Figure 2-1: Data0[18:10]={Control Bit, Data[7:0]}, Data0[8:0]
={Control Bit, Data[7:0]}, Data1[18:10] ={Control Bit, Data[7:0]}, Data1[8:0] ={Control Bit, Data[7:0]}.
TXDA_[19], TXDA_[9], TXDB_[19], and TXDB_[9] are unused, and should be grounded in this application
mode. See Appendix C for a full list of control characters supported in the 8b/10b encoder/decoder.
Symbol Transmission Order:
When 3.5/3.4 = 0: Data0[18:10] is the first encoded transmitted or decoded received symbol. Data0[8:0] is
next, followed by Data1[18:10].
When 3.5/3.4 = 1: Data0[8:0] is the first encoded transmitted or decoded received symbol. Data0[18:10] is
next, followed by Data1[8:0].
Bit Transmission Order within a Symbol:
Control Character Bits are always on TD*_[18], TD*_ [8], RD*_[18], RD*_[8]
Data bytes are always on TD*_[17:10], TD*_[7:0], RD*_[17:10], RD*_[7:0].
The most significant bit of the data byte is always on TD*_[17], TD*_[7], RD*_[17], RD*_[7], and is bit "H"
in Figure 2-2.
When 8.3/8.2 = 1, The "a" bit in Figure 2-2 is serially transmitted first or received first (typical case, shown
below) per symbol.
When 8.3/8.2 = 0, The "j" bit in Figure 2-2 is serially transmitted first or received first (atypical case,
reverse from order Figure 2-2) per symbol.
Figure 2-2. 16-bit SDR Parallel Interface Mode (Serial Bit Order)
20
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Figure 2-1 shows the two modes of operation in SDR mode, rising edge aligned mode and falling edge
aligned mode. In rising edge aligned mode, TDA_* and TDB_* inputs are sampled on the falling edges of
TXCLK_A and TXCLK_B respectively. In falling edge aligned mode, TDA_* and TDB_* inputs are
sampled on the rising edge of TXCLK_A and TXCLK_B respectively. In rising edge aligned mode, RDA_*
and RDB_* are timed such that an external device can sample the data using the falling edge of
RXCLK_A and RXCLK_B respectively. In falling edge aligned mode, RDA_* and RDB_* are timed such
that an external device can sample the data using the rising edge of RXCLK_A and RXCLK_B
respectively.
2.6
20-bit DDR (Double Data Rate) Mode (8b/10b Encoder/Decoder Disabled)
When DDR is enabled with the 8b/10b encoder disabled, the data format is identical to that of "20-bit SDR
(Single Data Rate) Mode (8b/10b Encoder/Decoder Disabled)" mode, except that four symbols are
transferred per parallel interface clock cycle instead of two. See the referenced previous section for further
details. Figure 2-3 shows the two modes of operation in DDR mode, source centered and source aligned
mode. In source centered mode, TDA_* and TDB_* inputs are sampled on the rising and falling edges of
TXCLK_A and TXCLK_B respectively. In source aligned mode, TDA_* and TDB_* inputs arrive
simultaneously with TXCLK_A and TXCLK_B respectively, and the TXCLK_A and TXCLK_B sampling
window is created internal to TLK6002 by delaying the clock. In source centered mode, RDA_* and RDB_*
are timed such that an external device can sample the data using RXCLK_A and RXCLK_B respectively,
where the appropriate timing window for sampling is created by TLK6002. In source aligned mode, RDA_*
and RDB_* are aligned with RXCLK_A and RXCLK_B respectively at the outputs of TLK6002, and the
sampling window must be created external to TLK6002.
DDR Source Centered Timing
TXCLK _A/B
TDA/B_[19:0]
Data 1[ 19:0]
Data 0[ 19:0]
RXCLK _A /B
RDA/B_[19:0]
Data 1[ 19:0]
Data 0[ 19:0]
DDR Source Aligned Timing
TXCLK _A/B
TDA/B_[19:0]
Data 0[ 19:0]
Data 1[ 19:0]
RXCLK _A /B
Data 0[ 19:0]
RDA/B_[19:0]
Data 1[ 19:0]
Figure 2-3. 20-bit DDR Parallel Interface Mode
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
21
TLK6002
SLLSE34 – MAY 2010
2.7
www.ti.com
16-bit DDR (Double Data Rate) Mode (8b/10b Encoder/Decoder Enabled)
When DDR is enabled with the 8b/10b encoder enabled, the data format is identical to that of "16-bit SDR
(Single Data Rate) Mode (8b/10b Encoder/Decoder Enabled)" mode, except that four symbols are
transferred per parallel interface clock cycle instead of two. See the referenced previous section for further
details.
2.8
Parallel Interface Clocking Modes
The TLK6002 supports source centered timing and source aligned DDR timing on the parallel receive
output bus. TLK6002 also supports rising edge aligned and falling edge aligned SDR timing on the parallel
receive output bus. See Figure 2-4 for more details.
RXCLK_A
tSETUP
Source Centered (DDR)
RDA_ [19:0 ]
tHOLD
tSETUP
tHOLD
Data
Data
Source Aligned (DDR)
RDA_ [19:0]
Data
Data
Data
Falling Edge Aligned (Rising Edge Sampled) (SDR)
RDA_ [19:0]
Data
Data
Rising Edge Aligned (Falling Edge Sampled ) (SDR)
RDA_ [19:0]
Data
Data
Figure 2-4. Receive Interface Timing – Source Centered/Aligned (Channel A is shown).
The transmit input timing modes are shown in Figure 2-5.
Transmit SDR/DDR input timing modes supported are similar to RX modes.
22
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
TXCLK _A
tSETUP
Source Centered (DDR)
TDA_ [19:0]
tHOLD
tSETUP
tHOLD
Data
Data
Source Aligned (DDR)
TDA_ [19:0]
Data
Data
Data
Falling Edge Aligned (Rising Edge Sampled ) (SDR)
TDA_[19:0]
Data
Data
Rising Edge Aligned (Falling Edge Sampled ) (SDR)
TDA_ [19:0]
Data
Data
Figure 2-5. Transmit Interface Timing (Channel A is shown).
2.9
Scrambler and De-scrambler
TLK6002 incorporates a scrambling function located before the 8b/10b encoder in the transmit datapath,
and a de-scrambling function located after the 8b/10b decoder in the receive datapath. The scrambler and
de-scrambler can be enabled/disabled using the MDIO management serial interface.
The transmitter applies a 7-degree polynomial to data bytes (not control), and the inverse operation is
performed by the receiver.
The scrambler/descrambler should be disabled if the 8b/10b encoder/decoder is disabled.
To achieve randomness between transmitting lanes, transmitters can be programmed to have differing
scrambling offset. Each transmitter seed value is programmed into a register which will be used by that
transmitter. The user should program unique seed values for adjacent TX links.
The receivers also have their own de-scrambling seed value registers. The receiver’s de-scrambling seed
value must be programmed to be the same as the corresponding transmitting end of the link. There is no
training sequence for transmitting the seed values to the receiver.
The scrambler is a 7-degree polynomial, linear feedback shift register (LFSR). The polynomial is; (X7 + X6
+ 1). K28.1, K28.5, or K28.7 characters reset the LFSR to the seed value. The bit pattern repeats every
127 bits.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
23
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Data In
Data Out
Sync Reset
Figure 2-6. 7-Degree Polynomial Scrambler
2.10 Power Down Mode
The TLK6002 can be put in power down either through device input pins or through MDIO control register
(1.15). PD_TRXA_N: Active low, powers down channel A. PD_TRXB_N: Active low, powers down channel
B.
The MDIO management serial interface remains operational when in register based power down mode
(1.15 asserted for both channels), but status bits may not be valid since the clocks are disabled. The serial
outputs and parallel output interface signals are high impedance when in power down mode. See the
detailed per pin description for behavior of each device I/O signal during pin based and register based
power down.
2.11 Parallel to Serial (Transmit):
In the transmit direction, the device accepts parallel input data on the TDA_[19:0] and TDB_[19:0] input
pins and converts the data into an optionally scrambled 8b/10b encoded serial stream on the TXAP/N and
TXBP/N serial output pins.
2.12 Serial to Parallel (Receive)
Serial data is received on the RXAP/N and RXBP/N pins, and optionally descrambled and 8b/10b
decoded and converted to parallel output data pins RDA_[19:0] and RDB_[19:0]. The interpolator and
clock recovery circuit will lock to the data stream if the incoming serial rate is within ±200 PPM of the
reference clock for the channel. The recovered byte clock is used to retime and deserialize the input data
stream, and is always synchronous with the parallel output data.
2.13 High Speed CML Output
The high speed data output driver is implemented using Current Mode Logic (CML) with integrated pull up
resistors, requiring no external components. The transmit outputs must be AC coupled.
TXAP
RXAP
50 W Transmission Line
50 W
0.8*VDDT
50 W
GND
50 W Transmission Line
TXAN
TRANSMITTER
RXAN
MEDIA
RECEIVER
Figure 2-7. Example of High Speed I/O AC Coupled Mode (Channel A is shown).
24
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Current Mode Logic (CML) drivers often require external components. The disadvantage of the external
component is a limited edge rate due to package and line parasitic. The CML driver on TLK6002 has
on-chip 50Ω termination resistors terminated to VDDT, providing optimum performance for increased
speed requirements. The transmitter output driver is highly configurable allowing output amplitude and
de-emphasis to be tuned to a channel's individual requirements. Software programmability allows for very
flexible output amplitude control. Only AC coupled output mode is supported.
When transmitting data across long lengths of PCB trace or cable, the high frequency content of the signal
is attenuated due to the skin effect of the media. This causes a "smearing" of the data eye when viewed
on an oscilloscope. The net result is reduced timing margins for the receiver and clock recovery circuits. In
order to provide equalization for the high frequency loss, 3-tap finite impulse response (FIR) transmit
de-emphasis is implemented. A highly configurable output driver maximizes flexibility in the end system by
allowing de-emphasis and output amplitude to be tuned to a channel’s individual requirements. Output
swing control is via MDIO.
See Figure 4-2 for output waveform flexibility. The level of de-emphasis is programmable via the MDIO
interface through control registers (2.12:4) through pre-cursor and post-cursor settings. Users can control
the strength of the de-emphasis to optimize for a specific system requirement.
2.14 High Speed Receiver
The high speed receiver is differential CML with internal termination resistors. The receiver requires AC
coupling. The termination impedances of the receivers are configured as 100Ω with the center tap weakly
tied to 0.8×VDDT with a capacitor to create an AC ground.
TLK6002 receiver incorporates an adaptive equalizer. This circuit compensates for channel insertion loss
by amplifying the high frequency components of the signal, reducing inter-symbol interference.
Equalization can be enabled or disabled per register settings. Both the gain and bandwidth of the
equalizer are controlled by the receiver equalization logic.
2.15 Loss Of Signal Output Signal Generation (LOS)
Loss of input signal detection is based on the voltage level of each serial input signal RXAP/N and
RXBP/N. Anytime the serial receive input differential signal peak to peak voltage level is ≤75 mVdfpp,
LOSA or LOSB are asserted (high true) respectively for Channel A and Channel B (if enabled, disabled by
default). Note that an input signal ≥ 150 mVdfpp is required for reliable operation of the loss of signal
detection circuit. If the input signal is between these two ranges, the SERDES will operate properly, but
the LOS indication will not be valid (or robust). The LOS indications are also directly readable through the
MDIO interface in register bits (5.2). The LOS indication per channel can be enabled through register bit
6.10 (defaults to disabled).
The following additional critical status conditions can be combined with the loss of signal condition
enabling additional realtime status signal visibility on the LOSA and LOSB outputs per channel:
1. GPI1 – Inverted and Logically OR'd (Register 6.11 enable) with LOS condition(s) when enabled – This
input signal, when enabled (disabled by default), is inverted and logically OR'd with the internally
generated LOS condition (on both channels) to allow easy overlay of additional board or external
device status with the other LOSA/LOSB indications.
2. Loss of Channel Synchronization Status – Logically OR'd with LOS condition(s) when enabled –
(Register 6.9 enabled). Loss of channel synchronization can be optionally logically OR'd (disabled by
default) with the internally generated LOS condition (per channel). In 20-bit operational mode, the
comma detection circuit must be enabled to actually enable this OR function. If it is not, this function is
not OR'd with the other LOS generating conditions. This bit should not be enabled unless comma
detection is enabled.
3. Loss of PLL Lock Status – Logically OR'd with LOS condition(s) when enabled – (Register 6.8
enabled). The internal PLL loss of lock status bit is optionally OR'd (disabled by default) with the other
internally generated loss of signal conditions (per channel).
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
25
TLK6002
SLLSE34 – MAY 2010
www.ti.com
4. Receive 8b/10b Decode Error (Invalid Code Word or Running Disparity Error) – Logically OR'd with
LOS condition(s) when enabled – (Register 6.3 enabled). The occurrence of an 8b/10b decode error
(invalid code word or disparity error) is optionally OR'd (disabled by default) with the other internally
generated loss of signal conditions (per channel).
5. ARS_Locked (ARS State Machine Currently Locked) – Inverted and Logically OR'd with LOS
condition(s) when enabled – (Register 10.14 enabled). ARS State Machine unlocked indication is
optionally OR'd (disabled by default) with the other internally generated loss of signal conditions (per
channel).
6. AGCLOCK (Active Gain Control Currently Locked) – Inverted and Logically OR'd with LOS
conditions(s) when enabled – (Register 7.7 enabled). RX SERDES adaptive gain control unlocked
indication is optionally OR'd (disabled by default) with the other internally generated loss of signal
conditions (per channel).
7. AZDONE (Auto Zero Calibration Done) – Inverted and Logically OR'd with LOS conditions(s) when
enabled – (Register 7.6 enabled). RX SERDES auto zero not done indication is optionally OR'd
(disabled by default) with the other internally generated loss of signal conditions (per channel).
See Figure 2-8, which shows the detailed implementation of the LOSA signal.
Loss of Signal (Ch A)
6.10 (Ch A)
GPI1
6.11 (Ch A)
Channel A In Sync
6.9 (Ch A)
PLL Locked (Ch A)
LOSA
6.8 (Ch A)
8b/10b Invalid Code (Ch A)
6.3 (Ch A)
ARS Locked (Ch A)
10.14 (Ch A)
AGCLOCK (Ch A)
7.7 (Ch A)
AZDONE (Ch A)
7.6 (Ch A)
NOTE: LOSA is asserted (driven high) during a failing condition, and deasserted (driven low) otherwise. Any combinations of
status signals may be enabled onto LOSA/B based on MDIO register bits indicated above. LOSB circuit is similar.
Figure 2-8. LOSA – Logic Circuit Implementation
26
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
2.16 Receive Datapath Error Condition Operation
The receive datapath (parallel output), when the 8b/10b decoder is enabled, automatically replaces the
received symbol with K30.7 (control = 1, data = 0xFE) in the case of an invalid code word or 8b/10b
disparity error.
The following additional conditions optionally enable replacement of received data with K30.7 (control = 1,
data = 0xFE) (when enabled through MDIO) when 8b/10b decoding is enabled, or replace the parallel
output data with all zero data if the 8b/10b decoder is disabled:
1. Loss of Signal Status – (character replacement enabled through register bit 6.6, disabled by default).
2. Loss of Channel Synchronization Status – (character replacement enabled through register bit 6.5,
disabled by default). This bit should not be enabled unless comma detection is enabled.
Note: Achieving channel synchronization is not possible if register bit 6.6 is high and LOS is detected.
3. Loss of PLL Lock Status – (character replacement enabled through register bit 6.4, disabled by default)
4. GPI1 – (if GPI1=0, character replacement enabled through register bit 6.7, disabled by default)
5. AZDONE – (if AZDONE=0, character replacement enabled through register bit 7.4, disabled by default)
6. AGCLOCK – (if AGCLOCK=0, character replacement enabled through register bit 7.5, disabled by
default)
2.17 Loopback Support
TLK6002 supports several loopback configurations.
Local loopback accepts parallel input data, and returns that data on the parallel output for the same
channel.
Remote loopback accepts serial input data, and returns that data on the serial output for the same
channel.
Shallow local loopback data traverses the entire transmit datapath except for serialization, and is returned
through the entire receive datapath (except for deserialization). Data is not serialized or deserialized.
Deep local loopback data traverses the entire transmit datapath including serialization, and is returned
through the entire receive datapath (including serialization). Data is both serialized and deserialized.
Deep remote loopback data traverses the entire receive datapath including the 20-bit output register, and
is returned through the entire transmit datapath (excluding the parallel input buffers). Data is both
deserialized and serialized.
Shallow remote loopback data traverses the entire receive datapath until just before the 20-bit output
register, and is returned through the entire transmit datapath (excluding the parallel input buffers). Data is
both serialized and deserialized.
Figure 2-9 and Figure 2-10 show all four loopback modes of operation.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
27
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Shallow Local Loopback ( register 7. 0)
RXCLK_A
8B/ 10B
Encoder
RDA_[19: 0]
20- bit
REGI STER
TXCLK_A
20- bit
REGI STER
Channel A
TDA_[19: 0]
COMMA
Detect &
8B/ 10 B
Decoding
TXAP
Parallel to
Serial
TXAN
RXAP
Serial to
Parallel
RXAN
Channel A
Deep Local Loopback ( register 7.1))
RXCLK_A
8B/ 10B
Encoder
R DA_[19: 0]
20- bit
REGI STER
TXCLK_A
20- bit
REGI STER
Channel A
TDA_[19: 0]
COMMA
Detect &
8B/ 10 B
Decoding
TXAP
Parallel to
Serial
TXAN
RXAP
Serial to
Parallel
RXAN
Channel A
Figure 2-9. TLK6002 Shallow and Deep Local Loopback (Channel A)
28
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Shallow Remote Loopback (register 7 .2 )
RXCLK_A
8B/ 10B
Encoder
RDA_[19: 0]
20- bit
REGI STER
TXCLK_A
20- bit
REGI STER
Channel A
TDA_[19: 0]
COMMA
Detect &
8B/ 10 B
Decoding
TXAP
Parallel to
Serial
TXAN
RXAP
Serial to
Parallel
RXAN
Channel A
Deep Remote Loopback (register 7 .3 )
RXCLK_A
8B/ 10B
Encoder
RDA_[ 19: 0]
20- bit
REGI STER
TXCLK_A
20- bit
REGI STER
Channel A
TDA_[19: 0]
COMMA
Detect &
8B/ 10 B
Decoding
TXAP
Parallel to
Serial
TXAN
RXAP
Serial to
Parallel
RXAN
Channel A
Figure 2-10. TLK6002 Shallow and Deep Remote Loopback (Channel A)
2.17.1 Link Test Functions
The TLK6002 has an extensive suite of built in test functions to support system diagnostic requirements.
Each channel has an internal test pattern generator and verifier. Several patterns can be selected via the
MDIO that offer extensive test coverage. The test patterns supported are: 27-1, 223-1, 231-1 PRBS
(Pseudo Random Bit Stream), CRPAT Short/Long frequency patterns.
2.18 Serial Retime Mode
TLK6002 supports serial retime mode of operation. Serial retime mode is enabled through an mdio
register bit.
In serial retime mode mode:
• Incoming serial data on RXAP/N is sent to both RDA_[19:0] parallel and TXBP/N serial outputs.
• Incoming serial data on RXBP/N is sent to both RDB_[19:0] parallel and TXAP/N serial outputs.
In serial retime mode, the incoming serial data rate on Channel A must be synchronous (0 ppm) to the
reference clock supplied to Channel B SERDES. Also, the incoming serial data rate on Channel B must be
synchronous (0 ppm) to the reference clock supplied to Channel A SERDES.
Note that latency measurement is not possible when in serial retime mode of operation.
Figure 2-11 shows operation of TLK6002 in serial retime mode:
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
29
TLK6002
SLLSE34 – MAY 2010
www.ti.com
RXCLK_A
8B/ 10B
Encoder
RDA_[19: 0]
20- bit
REGI STER
TXCLK_A
20- bit
REGI STER
Channel A
TDA_[19: 0]
COMMA
Detect &
8B/ 10 B
Decoding
TXAP
Parallel to
Serial
TXAN
RXAP
Serial to
Parallel
RXAN
Channel A
RXCLK_B
8B/ 10B
Encoder
RDB_[ 19: 0]
20- bit
REGI STER
TXCLK_B
20- bit
REGI STER
Channel B
TDB_[19: 0]
COMMA
Detect &
8B/ 10 B
Decoding
TXBP
Parallel to
Serial
TXBN
RXBP
Serial to
Parallel
RXBN
Channel B
Figure 2-11. TLK6002 – Serial Retime Mode of Operation
2.19 Latency Measurement Function
The TLK6002 includes a round trip latency measurement function to support CPRI and OBSAI base
station applications. The elapsed time from a comma (either encoded or unencoded) detected in the
transmit direction of a particular channel to a comma detected in the receive direction of the same channel
is measured and reported through the MDIO interface (TDA_[19:0] → RDA_[19:0] -or- TDB_[19:0] →
RDB_[19:0]). The function operates on one channel at a time. When 8b/10b encoding/decoding is
enabled, the following three control characters (containing commas) are monitored:
1. K28.1 (control = 1, data = 0x3C)
2. K28.5 (control = 1, data = 0xBC)
3. K28.7 (control = 1, data = 0xFC).
When 8b/10b encoding/decoding is disabled, the lower 7 bits of the Tx and Rx data stream are monitored
for either positive (7’b0011111) or negative (7’b1100000) comma characters.
Whether 8b/10b encoding/decoding is enabled or not, comma detection in the receive datapath must be
enabled (register bit 3.7).
The result of this measurement is readable through the MDIO interface through a 20-bit register. The
accuracy of the measurement is a function of the serial bit rate which the channel being measured is
operating at. The register will return a value of 0xFFFFF if the duration between transmit and receive
comma detection exceeds the depth of the counter. Only one measurement value is stored internally until
the 20-bit results counter is read. The counter will return zero in cases where a transmit comma was never
detected (indicating the results counter never began counting).
30
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
In full rate mode, the latency measurement function runs off of an internal clock which is equal to the
frequency of the transmit serial bit rate divided by four. In half rate mode, the latency measurement
function runs off of an internal clock which is equal to the serial bit rate divided by two. In quarter rate
mode, the latency measurement function runs off of an internal clock which is equal to the serial bit rate.
In eighth rate mode, the latency measurement function runs off of a clock which is equal to twice the serial
bit rate. The latency measurement accuracy in all cases is equal to plus or minus one latency
measurement clock period. The measurement clock can be divided down if a longer duration
measurement is required, in which case the accuracy of the measurement is accordingly reduced. The
high speed latency measurement clock is divided by either 1, 2, 4, or 8 via register 16.5:4. The
measurement clock used is always selected by the channel under test. The high speed latency
measurement clock may only be used when operating at one of the eight serial rates specified in the
CPRI/OBSAI specifications. It is also possible to run the latency measurement function off of the
recovered byte clock for the channel under test (and gives a latency measurement clock frequency equal
to the serial bit rate divided by 10) via register bit 16.2 (where the 16.5:4 divider value setting is ignored).
The accuracy for the standard based CPRI/OBSAI application rates is shown in Table 2-6, and assumes
the latency measurement clock is not divided down per user selection (division is required to measure a
duration greater than 682 µs). For each division of 2 in the measurement clock, the accuracy is also
reduced by a factor of two.
Table 2-6. CPRI/OBSAI Latency Measurement Function Accuracy
(Undivided Measurement Clock)
Gbps
Rate
Clock Frequency
(GHz)
Accuracy
(± ns)
0.6144
Eighth
1.2288
0.8138
0.768
Eighth
1.536
0.6510
1.2288
Quarter
1.2288
0.8138
1.536
Quarter
1.536
0.6510
2.4576
Half
1.2288
0.8138
3.072
Half
1.536
0.6510
4.9152
Full
1.2288
0.8138
6.144
Full
1.536
0.6510
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
31
TLK6002
SLLSE34 – MAY 2010
www.ti.com
The locations where the comma transmission and reception are measured in TLK6002 are shown in
Figure 2-12.
Pattern
Generator
10
2:1
MUX
Parallel to
Serial
10-bit
8B/10B
Encoder
20-bit
Scr ambler
10-bit
TX_FI FO
TDA_[19:0]
20-bit
REGI STER
10
TXAP
TXAN
10-bit
10-bit
MUX
¼, ½ &
Full rate
set
10-bit
TCLK
TXCLK _A
Selected
Reference
Clock
Clock
Synthesizer
Start
Counter
20-bit Latency
Counter
Pattern
Verifier
Stop
Counter
RXCLK _A
10-bit
De-Scr ambler
10-bit
RX_FI FO
RDA_[19:0]
20-bit
REGI STER
10-bit
COMMA
Detect&
8B/10B
Decoding
10-bit
Serial to
Parallel
EQ
RXAP
RXAN
Figure 2-12. Location of TX and RX Comma Character Detection (Only Channel A Shown)
2.20 CPRI/OBSAI Automatic Rate Sense (ARS) Function
An automatic rate sense (ARS) function is implemented in TLK6002 to facilitate determination of the
incoming CPRI/OBSAI serial link rate per channel.
When ARS is enabled, only three device input reference clock frequencies are supported: 122.88, 153.6,
and 307.2 MHz. ARS should not be enabled unless one of these three frequencies is available on either
REFCLK_0_P/N or REFCLK_1_P/N.
The ARS function per channel can operate off of either reference clock (REFCLK_0_P/N or
REFCLK_1_P/N), and is selected through device pins REFCLK_A_SEL for Channel A, and
REFCLK_B_SEL for Channel B (or alternatively mdio registers). The reference clock rate selection is
selected through channel A or B MDIO register bits (ARS_REF_FREQ[1:0], register bits 11.15:14), and
must be programmed for proper ARS operation (unless the default values matches the reference clock
input frequency).
32
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Using any one of the three supported reference clock frequencies allows all eight currently defined CPRI
and OBSAI rates to be achieved with a single reference clock frequency, and eliminates the need for
external hardware to support multiple frequencies for OBSAI and CPRI operation. See Table 2-7 for a list
of supported CPRI/OBSAI rates.
ARS can be enabled/disabled through device pins (RATE_A/B) or channel A or B MDIO register bits
(ARS_EN[1:0], register bits 10.13:12), Software control is enabled by setting RATE_A/RATE_B pins to
100. Pin control is enabled by setting RATE_A/B pins to 101/110/111. ARS can be enabled or disabled
independently on each channel.
ARS does not support determination of incoming serial rates other than the eight defined by the
CPRI/OBSAI specifications. ARS should not be enabled unless one of those incoming serial rates is
anticipated. See Table 2-7 for a list of the supported CPRI/OBSAI incoming serial rates.
When ARS is enabled, a state machine will continuously loop through (and override previously
programmed) relevant SERDES control settings for a given input reference clock frequency until either an
incoming serial bit rate is successfully determined (indicated by assertion of channel A or B MDIO register
ARS_LOCKED, register bit 5.10), or the ARS function is disabled (either through device pin or MDIO
software control). Note that the order attempted is always from the highest serial bit rate to the lowest
serial bit rate. There is an MDIO register enable per serial bit rate per channel (ARS_SBR_ENABLE[7:0],
register bit 11.13:6), such that any number between one and eight of the supported incoming serial rates
can be determined. If an incoming serial bit rate is successfully determined, and then subsequently lost,
the state machine will automatically continue searching for a stable rate (as long as ARS is not disabled),
first starting with the last successful rate (if existing, effecting a single retry of the last working rate), and
then continuing down sequentially through the enabled lower serial rates (or if there are none, starting
over with the highest enabled incoming serial rate settings).
The ARS function monitors the incoming 8b/10b encoded serial receive data, using both the comma
character and 8b/10b disparity errors for a given channel, to determine and validate the incoming serial
data rate. The channel synchronization state machine is implemented as specified in IEEE802.3-2002
Clause 36, Figure 36-9, Page 62. The channel synchronization state machine flowchart is shown in
Figure 2-14 Channel Synchronization Flowchart. The 8b/10b decoder is used in tandem with the channel
synchronization state machine to determine if rate sense is successful at a particular device setting. Note
that the 8b/10b decoder is used for the ARS function even if 8b/10b encoding/decoding is disabled for the
datapath of the channel. Parallel output data is always output in the pin/software selected format (i.e.,
unencoded or 8b/10b encoded or byte aligned), and is not a function of whether ARS is enabled. Also
note that the RX SERDES CDR lock indication (AGCLOCK) qualifies channel synchronization.
When an ARS enabled channel is found to be in the channel synchronization state, the following rate
settings (RATE_TX[1:0] register bits 1.7:6, RATE_RX[1:0] register bits 1.5:4, PLL_MULT[3:0] register bits
1.3:0) are available to be read through the MDIO interface, as indicated by the per channel ARS Locked
register bit (ARS_LOCKED register bit 5.10) being asserted high. If the ARS function is not currently
locked onto the incoming serial data, the ARS locked register bit (ARS_LOCKED register bit 5.10) will
read deasserted, and the rate settings are not valid (although they are always readable). It is also possible
through MDIO configuration to make the inverse of ARS_LOCKED indication visible on the LOSA/B
outputs per channel, and in this mode can be used as a software interrupt notification. After a successful
rate determination is made, the ARS function will continue to monitor channel synchronization status. If
channel synchronization is lost, the ARS state machine will begin looping through SERDES settings
(reattempting with the last working setting one time rather than with the next different setting) until either
ARS is disabled or channel synchronization is achieved. The ARS state machine will stay in a particular
setting (expected serial rate) attempting to achieve rate determination for the number of reference clock
cycles programmed in ARS_INTERVAL[20:0] (per channel register bits 11.4:0 / 12.15:0), which indicates a
duration of time defined as the number of reference clock periods times 1024. This register is sized such
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
33
TLK6002
SLLSE34 – MAY 2010
www.ti.com
that greater than 4.5 seconds of time can be programmed per attempted interval. Thus, the ARS state
machine will attempt to determine a particular serial rate for a programmable number of reference clock
periods where channel synchronization cannot be established before attempting the next (and different)
lower serial bit rate, in a repeating/looping fashion when the lowest enabled serial bit rate is attempted
unsuccessfully.
The ARS function overrides the following SERDES register settings (RATE_TX / RATE_RX / PLL_MULT).
MDIO writes to these registers do not impact the actual values controlling internal SERDES device
settings as long as ARS is enabled for that channel, and reads instead return the current settings (which
may or may not be valid) as controlled by the ARS state machine (and does so until ARS is disabled).
Table 2-7. ARS Looped Device Settings (Looping order highest → lowest enabled bit rate)
Reference Clock (MHz)
ARS Rate / Scale / Multiplier Settings Per Reference Clock
34
153.6
122.88
307.2
Standard
Serial Rate (Gbps)
Rate
Rate Scale
SERDES Multiplier Setting Loop
CPRI
0.6144
Eighth
4
16
20
8
10
OBSAI
0.768
Eighth
4
20
25
CPRI
1.2288
Quarter
2
16
20
8
OBSAI
1.536
Quarter
2
20
25
10
CPRI
2.4576
Half
1
16
20
8
CPRI/OBSAI
3.072
Half
1
20
25
10
CPRI
4.9152
Full
0.5
16
20
8
CPRI/OBSAI
6.144
Full
0.5
20
25
10
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
2.21 Clock Out Generation In ARS Mode (CLK_OUT_P/N)
Table 2-8 shows the CLK_OUT_P/N output clock frequency in each of the three reference clock frequencies.
Table 2-8. ARS CLK_OUT_P/N Frequencies
Selected Reference Clock (MHz)
ARS CLK_OUT_P/N Frequency Per Reference Clock
VCO Freq./2 (GHz)
Standard
Serial Rate (Gbps)
153.6
122.88
307.2
153.60
CLK_OUT Division from VCO Frequency/2
122.88
307.20
CLK_OUT_P/N Frequency (MHz)
CPRI
0.6144
1.2288
8
20
8
153.60
61.44
153.60
OBSAI
0.768
1.5360
10
25
10
153.60
61.44
153.60
CPRI
1.2288
1.2288
8
20
8
153.60
61.44
153.60
OBSAI
1.536
1.5360
10
25
10
153.60
61.44
153.60
CPRI
2.4576
1.2288
8
20
8
153.60
61.44
153.60
CPRI/OBSAI
3.072
1.5360
10
25
10
153.60
61.44
153.60
CPRI
4.9152
1.2288
8
20
8
153.60
61.44
153.60
CPRI/OBSAI
6.144
1.5360
10
25
10
153.60
61.44
153.60
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
35
TLK6002
SLLSE34 – MAY 2010
www.ti.com
If supplying reference clock through an external clock jitter cleaning device, the VCXO used with the
external cleaning device should be chosen such that REFCLK maintains ±100 ppm accuracy during ARS
rate determination, since CLK_OUT behavior will not be deterministic during changes between settings.
Per the above table, note that CLK_OUT is always a fixed frequency and attempts to remain synchronous
(0 ppm) to the incoming serial data rate.
If reference clock of 153.6 MHz is selected, and an external clock jitter cleaning device is used, the clock
cleaning device will need to be configured to multiply the output clock by 1x, since CLK_OUT_P/N is
153.6MHz.
If reference clock of 122.88 MHz is selected, and an external clock jitter cleaning device is used, the clock
cleaning device will need to be configured to multiply the output clock by 2x, since CLK_OUT_P/N is
61.44MHz.
If reference clock of 307.2 MHz is selected, and an external clock jitter cleaning device is used, the clock
cleaning device will need to be configured to multiply the output clock by 2x, since CLK_OUT_P/N is
153.6MHz.
Figure 2-13 shows the flow of the ARS state machine:
36
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
ARS Enabled
En?
Check Highest
Enabled Serial
Bit Rate
No
Yes
Start
Timeout
Counter
Change SERDES Settings
Disable ENRX
Drive ENPLL Low
>16?
No
PLL
Lock?
Yes
No
Wait for PLL Lock
TO?
Yes
Set
ARS_PLL_LOCK_ERR
Bit
PLL
Lock?
No
Yes
Enable ENRX &
Wait For AZDONE
>16 &
AZDONE
Yes
?
No
Automatic Datapath Reset
Check CH Synchronization
Yes
Check Next Lower
Enabled Serial Bit Rate
(If at lowest rate , go to
highest enabled rate )
TO?
Sync?
Yes
No
MDIO
Gate
No
TX FIFO and
RX FIFO
Auto Reset
Yes
Wait for TX FIFO RST
(ARS Locked)
Yes
ARS Locked
Try Last
Successful (Same)
Serial Bit Rate
One More Time Before
Continuing With Next
Lower Enabled Rate
Yes
Sync?
No
MDIO
FIFO RST
?
No
Legend:
TO = X*1024 REFCLK Period Counter Timeout
Sync = Channel Synchronization & (AGCLOCK |
!AGCLOCK_EN) & (!6.6 | !LOS)
>16 = Greater Than or Equal to 16 REFCLK Periods
Mdio Gate = ARS_MDIO_GATE
MDIO FIFO RST = Mdio Write to Set TX FIFO RST
Figure 2-13. ARS State Machine Flowchart
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
37
TLK6002
SLLSE34 – MAY 2010
www.ti.com
2.22 Transmit Serial Output During ARS Mode
The transmit serial output is always actively driven during ARS mode. The user has flexibility in the value
transmitted by the serial output during ARS. Note that since the PLL is shared between a TX and RX
channel, that the transmit serial rate will automatically follow the rate setting which ARS is validating in the
receive direction (whether it is subsequently determined to be correct or not).
The following bits impact transmitted serial output data:
1. ARS_TX_DATAPATH_OVERRIDE – ARS Transmit Datapath Override – (per channel Register bit
10.10) – When asserted, in tandem with register (ARS_TX_DATA[9:0], register bits 10.9:0), any fixed
or repeating sequence of 10 bits can be transmitted during ARS. When deasserted, the transmit
parallel interface input data is transmitted and serialized as received during ARS (and may not be
deterministic as the TX FIFO will collide on each rate change unless a fixed (static) pattern is input into
the parallel input interface making the fifo collision unimpacting to the datapath).
2. ARS_TX_MDIO_GATE – ARS Transmit MDIO Gate – per channel Register bit 10.11 – This bit is only
relevant if TX_DATAPATH_OVERRIDE is asserted. When this bit is deasserted, upon successful ARS
rate determination, the transmit datapath TX FIFO is automatically reset (centered) and continuity
between the parallel input data and serial output is established without MDIO interaction. When this bit
is asserted, the transmit datapath will not automatically switch over to serializing parallel input data at
the time the ARS state machine successfully validates the incoming serial data rate (although the TX
and RX FIFO are both automatically reset). This will give the opportunity for local MDIO firmware to
interactively manage any additional device settings. Specifically this gives the device interfacing to
TLK6002 the opportunity to read MDIO registers to determine the validated incoming serial rate,
manage any other device or system settings required, and also manage TXCLK_A/B synchronicity to
REFCLK at the proper data rate. After these steps are complete, the final step is to recenter the TX
FIFO (by manually issuing a TX FIFO reset (TXFIFO_RESET register bit 4.2). Transmit datapath
reliable operation is fully restored after the TX FIFO reset MDIO write transaction is completed, and
datapath continuity between parallel inputs and serial outputs is established.
At the time when ARS rate determination is successful, both a TX and RX FIFO reset is automatically
issued internal to TLK6002. Please note that if ARS_TX_MDIO_GATE is not asserted, there may be
difficulty in effectively recentering the transmit fifo. Anytime the TX FIFO collides, it automatically recenters
itself. This automatic recentering is triggered by the TXCLK_A/B and SERDES TX byte clock (multiplied
up and divided down REFCLK_A/B) being asynchronous or having excessive phase drift. The TX FIFO is
only effectively centered when the relationship between these two clocks has stabilized, at which point
issuing a TX FIFO reset (manual or automatic through collision) will optimally center the TX FIFO. Note
that careful external control of the TXCLK_A/B and REFCLK relationship (0 ppm and TXCLK_A/B at the
right data rate) must be managed for the mode where ARS_TX_MDIO_GATE is deasserted to work
reliably. If the clock relationship is still changing at the time of automatic recenter, the fifo may at some
point in the future need to automatically recenter itself (via collision), at which time the transmit serial data
will be briefly corrupted before resuming reliable operation. It is recommended that ARS_TX_MDIO_GATE
is asserted unless careful system operation has been analyzed.
In any ARS mode, note that the receive datapath software reset (not the same as RX FIFO reset) should
not be issued as channel synchronization will be lost, and ARS would inappropriately begin searching for
the incoming serial rate again (which is undesirable).
2.22.1 Receive Parallel Output Data During ARS Mode
The parallel outputs are always driven during ARS mode. During ARS mode, it is anticipated that channel
synchronization will typically remain lost during the rate determination process, and thus the parallel output
data will typically behave predictably as indicated in the previous paragraph labeled Receive Datapath
Error Condition Operation.
38
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
2.22.2 Receive Parallel Output Clock During ARS Mode
Per channel ARS_RX_CLK_EN (register bit 10.15) allows software programmability as to whether the
recovered output byte clock (RXCLK_A/B) is allowed to toggle (dynamically changing rates as the ARS
rate determination process executes), or if it is held fixed to zero until the incoming serial rate for the
channel has been determined. At the time rate determination is successful, the receive parallel output
interface clock is automatically allowed to toggle if it was prevented from toggling during ARS (and
enabling toggling is not delayed or gated by MDIO interaction in the case where ARS_TX_MDIO_GATE is
asserted).
Reset | (6.6 & LOS(Loss of Signal))
Loss Of Sync
(Enable Alignment)
Sync Status Not Ok
No Comma
Comma
Comma Detect 1
(Disable Alignment)
!Comma & !Invalid Decode
Invalid Decode
Comma
Comma Detect 2
!Comma & !Invalid Decode
Invalid Decode
Comma
Comma Detect 3
!Comma & !Invalid Decode
Invalid Decode
Note:
If CH_SYNC_HYSTERESIS[1:0] (3.12:11) is equal to 2'b00),
machine operates as drawn.
If CH_SYNC_HYSTERESIS[1:0] (3.12:11) is equal to 2'b01/
2’b10/2'b11, then a transition from all Sync Acquired states
occurs immediately upon detection of 1, 2, or 3 adjacent
invalid code words or disparity errors respectively.
Comma
A
Sync Acquired 1
(Sync Status Ok)
Invalid
B Decode
Sync Acquired 2
(good cgs = 0)
C
Invalid
Decode
Invalid Decode
Sync Acquired 3
(good cgs = 0)
Invalid
Decode
!Invalid
Decode
Invalid Decode
Sync Acquired 4
(good cgs = 0)
Invalid
Decode
!Invalid
Decode
!Invalid
Decode
Invalid Decode
Sync Acquired 2A
good cgs++
!invalid Decode &
A
good_cgs=3
Sync Acquired 3A
good cgs++
!invalid Decode &
B
good_cgs=3
Sync Acquired 4A
good cgs++
C
!invalid Decode&
good_cgs=3
!Invalid Decode&
good_cgs !=3
!Invalid Decode &
good_cgs !=3
!Invalid Decode &
good_cgs !=3
Figure 2-14. Channel Synchronization Flowchart
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
39
TLK6002
SLLSE34 – MAY 2010
www.ti.com
2.23 Output Clock Selection (CLK_OUT_P/N)
Table 2-9 details CLK_OUT_P/N as a function of device settings.
Table 2-9. CLK_OUT_P/N Frequencies (ARS Enabled and Disabled)
RATE_A[2:0]
RATE_B[2:0]
CLK_OUT_SEL
CLK_OUT_P/N
Selected Channel A
REFCLK Frequency (MHz)
Indicated By ARS_REF_FREQ
Selected Channel B
REFCLK Frequency (MHz)
Indicated By ARS_REF_FREQ
000/001/010/011
000/001/010/011
0
RXBCLK_A/(0.5:4)
x
x
000/001/010/011
000/001/010/011
1
RXBCLK_B/(0.5:4)
x
x
110
Not 110
x
153.6 MHz (ppm 0 to RXAP/N)
153.6
x
110
Not 110
x
61.44 MHz (ppm 0 to RXAP/N)
122.88
x
110
Not 110
x
153.6 MHz (ppm 0 to RXAP/N)
307.2
x
110
Not 110
x
61.44 MHz (ppm 0 to RXAP/N)
245.76
x
Not 110
110
x
153.6 MHz (ppm 0 to RXBP/N)
x
153.6
Not 110
110
x
61.44 MHz (ppm 0 to RXBP/N)
x
122.88
Not 110
110
x
153.6 MHz (ppm 0 to RXBP/N)
x
307.2
Not 110
110
x
61.44 MHz (ppm 0 to RXBP/N)
x
245.76
101
000/001/010/011
0
153.6 MHz (ppm 0 to RXAP/N)
153.6
x
101
000/001/010/011
0
61.44 MHz (ppm 0 to RXAP/N)
122.88
x
101
000/001/010/011
0
153.6 MHz (ppm 0 to RXAP/N)
307.2
x
101
000/001/010/011
0
61.44 MHz (ppm 0 to RXAP/N)
245.76
x
101
000/001/010/011
1
RXBCLK_B/(0.5:4)
x
x
000/001/010/011
101
1
153.6 MHz (ppm 0 to RXBP/N)
x
153.6
000/001/010/011
101
1
61.44 MHz (ppm 0 to RXBP/N)
x
122.88
000/001/010/011
101
1
153.6 MHz (ppm 0 to RXBP/N)
x
307.2
000/001/010/011
101
1
61.44 MHz (ppm 0 to RXBP/N)
x
245.76
000/001/010/011
101
0
RXBCLK_A/(0.5:4)
x
x
101
101
0
153.6 MHz (ppm 0 to RXAP/N)
153.6
x
101
101
0
61.44 MHz (ppm 0 to RXAP/N)
122.88
x
101
101
0
153.6 MHz (ppm 0 to RXAP/N)
307.2
x
101
101
0
61.44 MHz (ppm 0 to RXAP/N)
245.76
x
101
101
1
153.6 MHz (ppm 0 to RXBP/N)
x
153.6
101
101
1
61.44 MHz (ppm 0 to RXBP/N)
x
122.88
101
101
1
153.6 MHz (ppm 0 to RXBP/N)
x
307.2
101
101
1
61.44 MHz (ppm 0 to RXBP/N)
x
245.76
111
101/110
x
153.6 MHz (ppm 0 to RXBP/N)
x
153.6
111
101/110
x
61.44 MHz (ppm 0 to RXBP/N)
x
122.88
111
101/110
x
153.6 MHz (ppm 0 to RXBP/N)
x
307.2
111
101/110
x
61.44 MHz (ppm 0 to RXBP/N)
x
245.76
111
000/001/010/011/111
0
RXBCLK_A/(0.5:4)
x
x
111
000/001/010/011/111
1
RXBCLK_B/(0.5:4)
x
x
101/110
111
x
153.6 MHz (ppm 0 to RXAP/N)
153.6
x
101/110
111
x
61.44 MHz (ppm 0 to RXAP/N)
122.88
x
101/110
111
x
153.6 MHz (ppm 0 to RXAP/N)
307.2
x
101/110
111
x
61.44 MHz (ppm 0 to RXAP/N)
245.76
x
000/001/010/011/111
111
0
RXBCLK_A/(0.5:4)
x
x
000/001/010/011/111
111
1
RXBCLK_B/(0.5:4)
x
x
2.24 MDIO Management Interface
The TLK6002 supports the Management Data Input/Output (MDIO) Interface as defined in Clause 22 of
the IEEE 802.3 Ethernet specification. The MDIO allows register-based management and control of the
serial links. Normal operation of the TLK6002 is possible without use of this interface. However, some
features are accessible only through the MDIO.
The MDIO Management Interface consists of a bi-directional data path (MDIO) and a clock reference
(MDC). The port address is determined by control pins (see Table 2-10).
40
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 2-10. MDIO Related Signals
SIGNAL
TYPE
MDC
LVCMOS
1.5V/1.8V
Input VDDO3
MDIO
DESCRIPTION
Management Interface Clock. This clock is used to sample the MDIO signal.
LVCMOS
Management Interface Data. Bidirectional data line for MDIO Port is sampled on the rising edge of MDC.
1.5V/1.8V
THIS SIGNAL MUST BE EXTERNALLY PULLED UP TO VDDO3. Consult IEEE802.3 Clause 22 for an appropriate
Input/OutputVDDO3
resistance value.
PRTAD[4:0]
LVCMOS
1.5V/1.8V
Input
VDDO2/ VDDO1/
VDDO1/ VDDO1/
VDDO1
Port Address. Used to select the Port ID in Clause 22 MDIO mode.
PRTAD[4:1] selects a block of two sequential Clause 22 port addresses. Each channel is implemented as a different port
address, and can be accessed by setting the appropriate port address field within the Clause 22 MDIO transaction.
PRTAD[0] is not used functionally, but is needed for device testability with other devices in the family of products.
Channel A responds to port address 0 within the block of two port addresses.
Channel B responds to port address 1 within the block of two port addresses.
It is possible for 16 TLK6002 devices to logically share an MDIO bus.
In Clause 22, the top 4 control pins PRTAD[4:1] determine the device port address. In this mode the 2
individual channels in TLK6002 are classified as 2 different ports. So for any PRTAD[4:1] value there will
be 2 ports per TLK6002.
TLK6002 will respond if the 4 MSB's of PHY address field on MDIO protocol (PA[4:1]) matches
PRTAD[4:1]. The LSB of PHY address field (PA[0]) will determine which channel/port within TLK6002 to
respond to.
If PA[0] = 1b0, TLK6002 Channel A will respond.
If PA[0] = 1b1, TLK6002 Channel B will respond.
Write transactions which address an invalid register or device or a read only register will be ignored. Read
transactions which address an invalid register will return a 0.
MDIO Protocol Timing:
The Clause 22 timing required to read from the internal registers is shown in Figure 2-9. The Clause 22
timing required to write to the internal registers is shown in Figure 2-10.
MDC
0
MDIO
1
1
0
PA4
PA0
RA4
RA0
32 "1's"
Start
Preamble
(1)
Read
Code
PHY
Addr
REG
Addr
Pu1
D15
0
Turn
Around
D0
1
Data
Idle
Note that the 1 in the Turn Around section is externally pulled up, and driven to Z by TLK6002.
Figure 2-15. CL22 – Management Interface Read Timing
MDC
MDIO
0
1
32 "1's"
Preamble
Start
0
1
Write
Code
PA[4:0]
PHY
Addr
RA4
RA0
REG
Addr
1
0
Turn
Around
D15
D0
Data
1
Idle
Figure 2-16. CL22 – Management Interface Write Timing
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
41
TLK6002
SLLSE34 – MAY 2010
www.ti.com
The IEEE 802.3 Clause 22 specification defines many of the registers, and additional registers have been
implemented for expanded functionality.
Clause 22 Indirect Addressing:
The TLK6002 Register space is divided into two register groups. One register group can be addressed
directly through Clause 22, and one register group can be addressed indirectly through Clause 22. The
register group which can be addressed through Clause 22 indirectly is implemented in vendor specific
register space (16’h8000 onwards). Due to clause 22 register space limitations, an indirect addressing
method is implemented so that this extended register space can be accessed through clause 22. To
access this register space (16’h8000 onwards), an address control register (Reg 30, 5’h1E) should be
written with the register address followed by a read/write transaction to address data register (Reg 31,
5’h1F) to access the contents of the address specified in address control register.
Figure 2-17 and Figure 2-18 illustrate an example write transaction to Register 16’h8000 using indirect
addressing in Clause 22.
MDC
MDIO
0
1
0
32 "1's"
Write
Code
Start
Preamble
1
PA[4:0]
5'h1E
PHY
Addr
REG
Addr
1
0
Turn
Around
16'h8000
Data
1
Idle
Figure 2-17. CL22 – Indirect Address Method – Address Write
MDC
MDIO
0
1
0
32 "1's"
Write
Code
Start
Preamble
1
PA[4:0]
5'h1F
PHY
Addr
REG
Addr
1
0
DATA
Turn
Around
Data
1
Idle
Figure 2-18. CL22 – Indirect Address Method – Data Write
Figure 2-19 and Figure 2-20 illustrate an example read transaction to read contents of Register 16’h8000
using indirect addressing in Clause 22.
MDC
MDIO
0
1
32 "1's"
Preamble
Start
0
1
Write
Code
PA[4:0]
5'h1E
PHY
Addr
REG
Addr
1
0
Turn
Around
16'h8000
Data
1
Idle
Figure 2-19. CL22 – Indirect Address Method – Address Write
42
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
MDC
0
MDIO
1
1
0
PA4
PA
0
5'h1F
32 "1's"
Preamble
Start
Read
Code
PHY
Addr
REG
Addr
Pu1
0
Turn
Around
D15
D0
Data
1
Idle
Figure 2-20. CL22 - Indirect Address Method – Data Read
The IEEE 802.3 Clause 22 specification defines many of the registers, and additional registers have been
implemented for expanded functionality.
Description
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
43
TLK6002
SLLSE34 – MAY 2010
www.ti.com
3 PROGRAMMERS REFERENCE
The following registers can be addressed directly through Clause 22. Channel identification is based on
PHY (Port) address field. Registers 0x01- 0x0C, 0X14 are per channel basis.
Channel A can be accessed by setting LSB of PHY address to 0.
Channel B can be accessed by setting LSB of PHY address to 1.
Table 3-1. GLOBAL_CONTROL_1
BIT(s)
0.15
0.11
Address: 0x00
Default: 0x0600
NAME
DESCRIPTION
GLOBAL_RESET
GLOBAL_WRITE
0.10:8 HSTL_IMPED_CLK_DIV[2:
0]
0.7
HSTL_IMPED_CLK_SEL
ACCESS
Global reset (Channel A and B).
RW
SC (1)
0=
Normal operation (Default 1’b0)
1=
Resets TX and RX datapath including MDIO registers. Equivalent to asserting
RESET_N.
Global write enable.
RW
0=
Control settings written to Registers 0x01-0x0C, 0x14 are specific to channel
addressed (Default 1’b0)
1=
Control settings written to Registers 0x01-0x0C, 0x14 are applied to both Channel A
and Channel B regardless of channel addressed
HSTL Impedance Control clock divide selection.
000 = Divide by 1
001 = Divide by 2
010 = Divide by 4
011 = Divide by 8
100 = Divide by 16
101 = Divide by 32
110 = Divide by 64 (Default 3’b110)
111 = Divide by 128
This value, in tandem with register bit 0.7, selects the frequency and source of the clock
used for dynamic voltage, temperature, and impedance compensation in the HSTL I/O
buffers. The division value selected should yield a value less than 10 MHz, and is
calculated by dividing the selected REFCLK_0/1_P/N frequency by the value above.
RW
HSTL Impedance Control reference clock source selection.
RW
0=
Selects channel A reference clock (as selected by REFCLK_A_SEL) as clock
reference to HSTL impedance control (Default 1’b0)
1=
Selects channel B reference clock (as selected by REFCLK_B_SEL) as clock
reference to HSTL impedance control
See register bit 0.10:8 for further details.
0.6
CLKOUT_SEL
Output clock select. Selected RXBCLK_A/B or ARS output clock is sent out on
CLK_OUT_P/N pins . Logically OR’ed with CLK_OUT_SEL pin.
0=
Selects Channel A recovered byte clock (RXBCLK_A) as output clock (Default 1’b0)
1=
Selects Channel B recovered byte clock (RXBCLK_B) as output clock.
RW
See Figure 1-3
0.5:4
CLKOUT_DIV[1:0]
Output clock divide setting in non-ARS mode.This value is used to divide selected
RXBCLK before giving it out onto CLK_OUT_P/N.
CLK_OUT_P/N Frequency = (Serial Bit Rate / 10)/(Register 0.5:4 Setting)
00
01
10
11
= Divide by
= Divide by
= Divide by
= Divide by
RW
1 (Default 2’b00)
2
4
8
See Table 2-9 and Figure 1-3
(1)
44
After reset bit is set to one, it automatically sets itself back to zero on the next MDC clock cycle.
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-1. GLOBAL_CONTROL_1 (continued)
Address: 0x00
BIT(s)
0.2
Default: 0x0600
NAME
DESCRIPTION
RETIME_EN
ACCESS
In this mode, serial A input data is sent to both the Parallel A output and Serial B output
interface. Serial B input data is sent to both the Parallel B output and Serial A output
interface.
0=
Normal functional mode (Default 1’b0)
1=
Enable retime mode
RW
Serial A input data rate must match (0 ppm) channel B reference clock.
Serial B input data rate must match (0 ppm) channel A reference clock
See Figure 2-11
0.1
REFCLK_A_SEL
Channel A Reference clock selection. Logically OR’ed with REFCLK_A_SEL pin.
0=
Selects REFCLK_0_P/N as clock reference to Channel A serdes macro (Default
1’b0)
1=
Selects REFCLK_1_P/N as clock reference to Channel A serdes macro
RW
See Figure 1-3
0.0
REFCLK_B_SEL
Channel B Reference clock selection. Logically OR’ed with REFCLK_B_SEL pin.
0=
Selects REFCLK_0_P/N as clock reference to Channel B serdes macro (Default
1’b0)
1=
Selects REFCLK_1_P/N as clock reference to Channel B serdes macro
RW
See Figure 1-3
Table 3-2. CHANNEL_CONTROL_1
Address: 0x01
BIT(s)
1.15
Default: 0x010D
NAME
POWERDOWN
DESCRIPTION
0=
Normal operation (Default 1’b0)
1=
Power Down mode is enabled.
1.10
RESERVED
For TI use only.
1.9:8
LOOP_BANDWIDTH
Serdes PLL Loop Bandwidth settings
1.7:6
RATE_TX [1:0]
ACCESS
Setting this bit high powers down the SERDES datapath channel with exception that MDIO
interface stays active. Logically OR’ed with inverse of PD_TRXx_N.
RW
00 =
Reserved
01 =
Applicable when external JC_PLL is NOT used (Default 2’b01)
10 =
Applicable when external JC_PLL is used
11 =
Reserved
Serdes TX rate settings
00 =
Full rate (Default 2’b00)
01 =
Half rate
10 =
Quarter rate
11 =
Eighth rate
RW
RW
Values written are valid only when RATE_A/B[2:0] pins are set to 3’b100 and ARS is not
enabled (10.13:12 = 2’b00)
In ARS mode, values written to these bits are not considered to determine serdes TX rate.
Instead it is automatically determined by ARS machine.
When read, these bits will reflect the final serdes Tx rate values
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
45
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-2. CHANNEL_CONTROL_1 (continued)
Address: 0x01
BIT(s)
1.5:4
Default: 0x010D
NAME
DESCRIPTION
RATE_RX [1:0]
ACCESS
Serdes RX rate settings
00 =
Full rate (Default 2’b00)
01 =
Half rate
10 =
Quarter rate
11 =
Eighth rate
RW
Values written are valid only when RATE_A/B[2:0] pins are set to 3’b100 and ARS is not
enabled (10.13:12 = 2’b00)
In ARS mode, values written to these bits are not considered to determine serdes RX rate.
Instead it is automatically determined by ARS machine.
When read, these bits will reflect the final serdes Rx rate values
1.3:0
PLL_MULT[3:0]
Serdes PLL multiplier setting (Default 4’b1101). In ARS mode, values written to these bits is
not considered to determine serdes PLL multiplier. Instead it is automatically determined by
ARS machine and these bits when read will reflect that value.
RW
Refer Table 3-3
See Appendix B for more information on PLL multiplier mettings
Table 3-3. PLL Multiplier Control
1[3:0]
1[3:0]
VALUE
PLL MULTIPLIER FACTOR
VALUE
PLL MULTIPLIER FACTOR
0000
Reserved
1000
12x
0001
Reserved
1001
Reserved
0010
4x
1010
15x
0011
5x
1011
16x
0100
6x
1100
16.5x
0101
8x
1101
20x
0110
8.25x
1110
25x
0111
10x
1111
Reserved
Table 3-4. CHANNEL_CONTROL_2
Address: 0x02
Default: 0x000A
Bit(s)
Name
Description
2.12:8
TWPOST1[4:0]
Adjacent post cursor Tap weight. Selects TAP settings for TX waveform. (Default 5’b00000)
Refer Table 3-5
RW
2.7:4
TWPRE[3:0]
Pre cursor Tap weight. Selects TAP settings for TX waveform. (Default 4’b0000)
Refer Table 3-6
RW
2.3:0
SWING[3:0]
Transmitter Output swing control for Serdes. (Default 4’b1010)
Refer Table 3-7
RW
46
Access
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-5. Post-Cursor Transmit Tap Weights
2[12:8]
2[12:8]
VALUE
TAP WEIGHT (%)
VALUE
TAP WEIGHT (%)
00000
0
10000
0
00001
+2.5
10001
–2.5
00010
+5.0
10010
–5.0
00011
+7.5
10011
–7.5
00100
+10.0
10100
–10.0
00101
+12.5
10101
–12.5
00110
+15.0
10110
–15.0
00111
+17.5
10111
–17.5
01000
+20.0
11000
–20.0
01001
+22.5
11001
–22.5
01010
+25.0
11010
–25.0
01011
+27.5
11011
–27.5
01100
+30.0
11100
–30.0
01101
+32.5
11101
–32.5
01110
+35.0
11110
–35.0
01111
+37.5
11111
–37.5
Table 3-6. Pre-Cursor Transmit Tap Weights
2[7:4]
2[7:4]
VALUE
TAP WEIGHT (%)
VALUE
TAP WEIGHT (%)
0000
0
1000
0
0001
+2.5
1001
–2.5
0010
+5.0
1010
–5.0
0011
+7.5
1011
–7.5
0100
+10.0
1100
–10.0
0101
+12.5
1101
–12.5
0110
+15.0
1110
–15.0
0111
+17.5
1111
–17.5
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
47
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-7. AC Mode Output Swing Control
VALUE
2[3:0]
AC MODETYPICAL AMPLITUDE
(mVdfpp)
0000
126
0001
215
0010
303
0011
395
0100
478
0101
572
0110
662
0111
756
1000
839
1001
932
1010
1020
1011
1110
1100
1190
1101
1280
1110
1360
1111
1450
Table 3-8. CHANNEL_CONTROL_3
Address: 0x03
BIT(s)
3.12:11 CH_SYNC_HYSTERESIS
[1:0]
3.10
Default: 0x0180
NAME
TX_SWAP_SEL
DESCRIPTION
ACCESS
Valid only when comma detection (channel synchronization) is enabled.
00 =
The channel synchronization, when in the synchronization state, performs the
Ethernet standard specified hysteresis to return to the unsynchronized state
(Default 2’b00)
01 =
A single 8b/10b invalid decode error or disparity error causes the channel
synchronization state machine to immediately transition from sync to unsync
10 =
Two adjacent 8b/10b invalid decode errors or disparity errors cause the channel
synchronization state machine to immediately transition from sync to unsync
11 =
Three adjacent 8b/10b invalid decode errors or disparity errors cause the
channel synchronization state machine to immediately transition from sync to
unsync
0=
Selects same channel Tx parallel interface input data as core input data for that
channel (Default 1’b0)
1=
Selects partner channel Tx parallel interface input data as core input data for
that channel
RW
RW
See Figure 1-4
3.9
RX_SWAP_SEL
0=
Selects same channel deserialized input data to be sent out the Rx parallel
output interface (Default 1’b0)
1=
Selects partner channel deserialized input data to be sent out the Rx parallel
output interface
RW
See Figure 1-5
3.8
RXCLK_OUT_SEL
Parallel output clock (RXCLK_x) selection
RW
When RX_SWAP_SEL (3.9) is 1’b0
0=
Selects respective channel SERDES TXBCLK clock
1=
Selects respective channel SERDES RXBCLK clock (Default 1’b1)
When RX_SWAP_SEL (3.9) is 1’b1
0=
Selects partner channel SERDES TXBCLK clock
1=
Selects partner channel SERDES RXBCLK clock (Default 1’b1)
See Figure 1-5
48
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-8. CHANNEL_CONTROL_3 (continued)
Address: 0x03
BIT(s)
3.7
Default: 0x0180
NAME
COMMA_ENABLE
DESCRIPTION
0=
Disables comma detection
1=
Enables comma detection (Default 1’b1)
ACCESS
RW
Comma detection automatically enabled during CRPAT verification.
3.6
3.5
3.4
3.3
DDR_ENABLE
TX_SYMBOL_ORDER
RX_SYMBOL_ORDER
ENCODE_ENABLE
0=
Enables SDR data mode on parallel Transmit and Receive directions (data is
clocked only on rising edge or only on falling edge) (Default 1’b0)
1=
Enables DDR data mode on parallel Transmit and Receive directions (data
clocked on both rising and falling edge)
0=
TDx_[19:10] symbol is serialized before TDx_[9:0]. (Default 1’b0)
1=
TDx_[9:0] symbol is serialized before TDx_[19:10].
0=
RDx_[19:10] symbol is deserialized before RDx_[9:0] symbol. (Default 1’b0)
1=
RDx_[9:0] symbol is deserialized before RDx_[19:10].
Encoder enable control. Logically OR’ed with CODE*_EN pin.
0=
8B/10B encode function is disabled (Default 1’b0)
1=
8B/10B encode function is enabled
RW
RW
RW
RW
Encoder automatically enabled during CRPAT test pattern generation.
3.2
DECODE_ENABLE
Decoder enable control. Logically OR’ed with CODE*_EN pin.
0=
8B/10B decode function is disabled (Default 1’b0)
1=
8B/10B decode function is enabled
RW
Decoder automatically enabled during CRPAT verification.
3.1
TX_EDGE_MODE
Transmit parallel input interface mode select. (Default 1’b0)
RW
When channel is in DDR mode
0=
Source centered timing on transmit parallel interface. Data is sampled on both
rising and falling clock edges.
1=
Source aligned timing on transmit parallel interface. Data is aligned with both
rising and falling clock edges, and a sampling point is created internal to
TLK6002 .
When channel is in SDR mode
3.0
RX_EDGE_MODE
0=
Falling edge align mode. Incoming data is aligned to falling edge of parallel
input clock. Internally data is sampled at the rising edge of the clock
1=
Rising edge align mode. Incoming parallel data is aligned to rising edge of
parallel input clock. Internally data is sampled at the falling edge of the clock.
Receive Parallel output interface mode select. (Default 1’b0)
RW
When channel is in DDR mode:
0=
Source centered timing on receive parallel interface. Data changing is offset
from the rising and falling clock edge per the HSTL Timing specification section,
and is easily sampled by external devices.
1=
Source aligned timing on receive parallel interface. Data changes at clock edge,
and a sampling point must be created by external devices.
When channel is in SDR mode:
0=
Falling edge align mode. Outgoing parallel data is aligned to the falling edge of
the parallel output clock, and is sampled on the rising edge by external devices.
1=
Rising edge align mode. Outgoing parallel data is aligned to the rising edge of
the parallel output clock, and is sampled on the falling edge by external devices.
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
49
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-9. CHANNEL_CONTROL_4
Address: 0x04
BIT(s)
4.3
Default: 0x0000
NAME
DATAPATH_RESET
DESCRIPTION
ACCESS
Channel datapath reset control. Required once the desired functional mode is configured.
RW
SC (1)
0 = Normal operation. (Default 1’b0)
1 = Resets channel logic excluding MDIO registers. (Resets both Tx and Rx datapath)
4.2
TXFIFO_RESET
Transmit FIFO reset control
0 = Normal operation. (Default 1’b0)
1 = Resets transmit datapath FIFO.
4.1
RXFIFO_RESET
Receive FIFO reset control
0 = Normal operation. (Default 1’b0)
1 = Resets receive datapath FIFO.
(1)
After reset bit is set to one, it automatically sets itself back to zero on the next MDC clock cycle.
Table 3-10. CHANNEL_STATUS_1
Address: 0x05
BIT(s)
5.14
Default: 0x0000
NAME
AZ_DONE
DESCRIPTION
ACCESS
Auto zero complete indicator.
RO/LL
When high, indicates auto zero calibration is complete
5.13
AGC_LOCKED
5.12
TP_ STATUS
Adaptive gain control loop lock indicator.
When high, indicates AGC loop is in locked state
Test Pattern status for test pattern selected in 7.10:8 .
RO
0 = Alignment has not achieved
1 = Alignment has been determined and correct pattern has been received.
Any bit errors received are reflected in ERROR_COUNTER register (0x0E)
5.11
ARS_PLL_LOCK_ERROR
ARS PLL Lock error indicator. Valid only when ARS function is enabled.
RO/LH
When high, indicates ARS did not detect PLL lock within the time specified through
ARS_INTERVAL[20:0] (11.4:0,12.15:0).
5.10
ARS_LOCKED
5.9
ENCODE_INVALID
ARS lock indicator. Valid only when ARS function is enabled.
RO/LL
When high, indicates ARS has determined incoming serial data rate.
Valid in 16 bit (SDR/DDR) mode (encoder enabled) and during
RO/LH
CRPAT test pattern generation.
When high, indicates encoder received an invalid control word.
5.8
DECODE_INVALID
Valid in 16 bit (SDR/DDR) mode (decoder enabled) and during CRPAT test pattern
verification.
RO/LH
When high, indicates decoder received an invalid code word, or a 8b/10b disparity error.
In functional mode, number of DECODE_INVALID errors are reflected in
ERROR_COUNTER register (0x0E)
5.7
TX_FIFO_UNDERFLOW
When high, indicates underflow has occurred in the transmit datapath FIFO.
RO/LH
5.6
TX_FIFO_OVERFLOW
When high, indicates overflow has occurred in the transmit datapath FIFO.
RO/LH
5.5
RX_FIFO_UNDERFLOW
When high, indicates underflow has occurred in the receive datapath FIFO.
RO/LH
5.4
RX_FIFO_OVERFLOW
When high, indicates overflow has occurred in the receive datapath FIFO.
RO/LH
5.3
GPI1
GPI1 status Indicator.
Reflects GPI1 input signal status.
5.2
LOS
Loss of Signal Indicator.
RO
RO/LH
When high, indicates that a loss of signal condition is detected on serial receive inputs
5.1
CHANNEL_SYNC
Channel synchronization status indicator. Valid only when comma detection is enabled
RO/LL
When high, indicates channel synchronization has achieved
5.0
PLL_LOCK
Serdes PLL lock indicator
RO/LL
When high, indicates Serdes PLL is locked to the selected incoming REFCLK_0/1_P/N
50
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-11. OVERRIDE_CONTROL
Address: 0x06
BIT(s)
6.15
6.14:12
Default: 0xC000
NAME
DESCRIPTION
ACCESS
RXCLK_OUT_EN
0 = Holds parallel output clock RXCLK_x output fixed at zero.
1 = Allows RXCLK_x output to toggle normally. (Default 1’b1)
RW
TX_FIFO_DEPTH[2:0]
TX FIFO Latency Control (Default 3’b100) This selection allows TX FIFO crash immunity to
be traded off against datapath latency. Selecting a large latency allows for the most dynamic
phase variation between TXCLK_A/B and the selected SERDES channel reference clock
(REFCLK_0/1_P/N). Careful consideration should be made in selecting this value based on
the anticipated phase movement between TXCLK_A/B and selected REFCLK_0/1_P/N
during system operation to avoid unwanted reoccurring transmit FIFO collision.
RW
TX_FIFO_DEPTH
[2:0]
TXCLK_A/B and selected
REFCLK_0/1_P/N. Relative
Phase Movement Allowed
(Serial Bit Times)
Appendix D Datapath Latency
Maximum Value (Serial Bit
Times)
000
±4
Same
001
±14
Same +20
010
±34
Same +60
011
±54
Same +110
1xx (Auto Selection) Full Rate: Same as 011
Half Rate: Same as 010
Quarter Rate: Same as 001
Eighth Rate: Same as 000
Consistent with above four
selections.
6.11
GPI_OVERLAY
0 = LOSx pin does not reflect GPI1 input signal status (Default 1’b0)
1 = Allows inverse value of GPI1 input signal to be reflected on LOSx pin
RW
6.10
LOS_OVERLAY
0 = LOSx pin does not reflect Serdes Rx Loss of signal condition (Default 1’b0)
1 = Allows Serdes Rx Loss of signal condition to be reflected on LOSx pin
RW
6.9
CH_SYNC_OVERLAY
0 = LOSx pin does not reflect loss of channel synchronization status (Default 1’b0)
1 = Allows channel loss of synchronization to be reflected on LOSx pin
RW
6.8
PLL_LOCK_OVERLAY
0 = LOSx pin does not reflect loss of PLL lock status (Default 1’b0)
1 = Allows loss of PLL lock status to be reflected on LOSx pin
RW
6.7
RX_CHAR_CTRL_GPI1
Receive data replacement control when GPI1 input is low
0 = Data passed through as received (Default 1’b0)
1 = Data replaced with K30.7 (when decoder is enabled) or all 0’s (when decoder is disabled)
RW
6.6
RX_CHAR_CTRL_LOS
Receive data replacement control during Loss of signal condition
0 = Data passed through as received (Default 1’b0)
1 = Data replaced with K30.7 (when decoder is enabled) or all 0’s (when decoder is disabled)
RW
6.5
RX_CHAR_CTRL_CH_SYNC
Receive data replacement control during Loss of synchronization condition
0 = Data passed through as received (Default 1’b0)
1 = Data replaced with K30.7 (when decoder is enabled) or all 0’s (when decoder is disabled)
RW
6.4
RX_CHAR_CTRL_PLL_LOCK
Receive data replacement control during Loss of PLL Lock condition
0 = Data passed through as received (Default 1’b0)
1 = Data replaced with K30.7 (when decoder is enabled) or all 0’s (when decoder is disabled)
RW
6.3
INVALID_CODE_OVERLAY
0 = LOSx pin does not reflect invalid code word error (Default 1’b0)
1 = Allows invalid code word error to be reflected on LOSx pin
RW
6.2
HSTL_SLEW_RATE
Slew Rate setting for RX parallel outputs
0 = No slew control (fastest edge) (Default 1’b0)
1 = 33% slower slew control
RW
HSTL_TERM[1:0]
Parallel Input Termination setting for TX parallel inputs
00 = Termination disable (High Impedance) (Default 2’b00)
01 = Half termination strength (200 Ω to VHSTL and GND) – Thevenin equivalent of 100 Ω to
(VDDQA/B)/2
1x = Full termination strength (100 Ω to VHSTL and GND) – Thevenin equivalent of 50 Ω to
(VDDQA/B)/2.
RW
6.1:0
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
51
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-12. LOOPBACK_TP_CONTROL
Address: 0x07
BIT(s)
7.15:14 RESERVED
DESCRIPTION
ACCESS
For TI use only.
RW
7.13
TP_GEN_EN
0 = Normal operation (Default 1’b0)
1 = Activates test pattern generation selected by bits 7.10:8
RW
7.12
TP_VERIFY_EN
0 = Normal operation (Default 1’b0)
1 = Activates test pattern verification selected by bits 7.10:8
RW
TEST_PATTERN_SEL[2:0]
Test Pattern Selection
000 = Reserved
001 = Reserved
010 = Reserved
011 = CRPAT Long (Supported only in half/quarter/eighth rate modes).
100 = CRPAT Short (Supported only in half/quarter/eighth rate modes).
101 = 27 - 1 PRBS pattern
110 = 223 - 1 PRBS pattern
111 = 231 - 1 PRBS pattern (Default 3’b111)
RW
7.7
AGCLOCK_OVERLAY
0 = LOSx pin does not reflect AGC unlock status (Default 1’b0)
1 = Allows AGC unlock status to be reflected on LOSx pin
RW
7.6
AZDONE_OVERLAY
0 = LOSx pin does not reflect auto zero calibration not done status (Default 1’b0)
1 = Allows auto zero calibration not done status to be reflected on LOSx pin
RW
7.5
RX_CHAR_CTRL_AGCLOCK
Receive data replacement control during AGC unlock condition
0 = Data passed through as received (Default 1’b0)
1 = Data replaced with K30.7 (when decoder is enabled) or all 0’s (when decoder is
disabled)
RW
7.4
RX_CHAR_CTRL_AZDONE
Receive data replacement control during auto zero calibration not done condition
0 = Data passed through as received (Default 1’b0)
1 = Data replaced with K30.7 (when decoder is enabled) or all 0’s (when decoder is
disabled)
RW
7.3
DEEP_REMOTE_LPBK
In this mode, serial input data traverses entire Rx datapath just before parallel
output drivers and is returned through entire Tx datapath and to the channel’s serial
output. The serial input data will also be available at the parallel output port. Serial
input data rate must match (0 ppm) reference clock.
0 = Normal functional mode (Default 1’b0)
1 = Enable deep remote loopback mode
See Figure 2-10
RW
7.2
SHALLOW_REMOTE_LPBK
In this mode, serial input following decoder is fed back to the channel’s encoder
input and to the serial output. The serial input data will also be available at the
parallel output port. Serial input data rate must match (0 ppm) reference clock
0 = Normal functional mode (Default 1’b0)
1 = Enable shallow remote loopback mode
See Figure 2-10
RW
7.1
DEEP_LOCAL_LPBK
In this mode, parallel input is looped after serializer and fed back to the channel’s
parallel output. The parallel input data is not available at the TX serial output. Note:
SWING[3:0] must be set to 4’b0000 in this mode.
0 = Normal functional mode (Default 1’b0)
1 = Enable deep local loopback mode
See figure Figure 2-9
Note that AZDONE must be asserted before asserting DEEP_LOCAL_LPBK. Also
note that AGCLOCK is not valid during DEEP_LOCAL_LPBK.
RW
7.0
SHALLOW_LOCAL_LPBK
In this mode, parallel input after the encoder and before the serializer is fed back to
the channel’s parallel output. The parallel input data will also be serialized and
available at the TX serial port
0 = Normal functional mode (Default 1’b0)
1 = Enable shallow local loopback mode
See Figure 2-9
RW
7.10:8
52
Default: 0x0700
NAME
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-13. SERDES_CONTROL_1
Address: 0x08
BIT(s)
Default: 0x3D4C
NAME
DESCRIPTION
ACCESS
8.14:12 EQPRE[2:0]
Serdes Rx precursor equalizer selection
000 = 1/9 cursor amplitude
001 = 3/9 cursor amplitude
010 = 5/9 cursor amplitude
011 = 7/9 cursor amplitude (Default 3’b011)
100 = 9/9 cursor amplitude
101 = 11/9 cursor amplitude
110 = 13/9 cursor amplitude
111 = Disable
RW
8.11:10 CDRTHR[1:0]
Clock data recovery algorithm threshold selection
00 = Four vote threshold
01 = Eight vote threshold
10 = Sixteen vote threshold
11 = Thirty two vote threshold (Default 2’b11)
RW
8.9:8
CDRFMULT[1:0]
Clock data recovery algorithm frequency multiplication selection
00 = First order. Frequency offset tracking disabled
01 = Second order. 1x mode (Default 2’b01)
10 = Second order. 2x mode
11 = Reserved
RW
8.7:6
AGCCTRL[1:0]
Adaptive gain control loop
00 = Attenuator will not change after lock has been achieved, even if AGC becomes
unlocked
01 = Attenuator will not change when in lock state, but could change when AGC becomes
unlocked (Default 2’b01)
10 = Force the attenuator off.
11 = Force the attenuator on
RW
8.5:4
AZCAL[1:0]
Auto zero calibration.
00 = Auto zero calibration initiated when receiver is enabled (Default 2’b00)
01 = Auto zero calibration disabled
10 = Forced with automatic update.
11 = Forced without automatic update
RW
8.3
TX_PMA_BIT_ORDER Determines whether LSB or MSB of the parallel inputs to be sent out first on serial
transmit pins.
0 = TDx[19] or TDx[9] (MSB) is serially transmitted first
1 = TDx[10] or TDx[0] (LSB) is serially transmitted first (Default 1’b1)
RW
8.2
RX_PMA_BIT_ORDER Determines whether serially received first bit to be sent out on LSB or MSB of the parallel
outputs.
0 = RDx[19] or RDx[9] (MSB) is serially received first
1 = RDx[10] or RDx[0] (LSB) is serially received first (Default 1’b1)
RW
8.1
TX_INVPAIR
Transmitter polarity.
0 = Normal polarity. TXxP considered positive data and TXxN considered negative data
(Default 1’b0)
1 = Inverted polarity. TXxP considered negative data and TXxN considered positive data
RW
8.0
RX_INVPAIR
Receiver polarity.
0 = Normal polarity. RXxP considered positive data. RXxN considered negative data
(Default 1’b0)
1 = Inverted polarity. RXxP considered negative data. RXxN considered positive data
RW
Table 3-14. SCRAMBLER_CONTROL
Address: 0x09
BIT(s)
9.15
NAME
TX_SCRAMBLER_EN
9.6:0
DESCRIPTION
ACCESS
0 = Disables scrambler on the transmit datapath (Default 1’b0)
1 = Enables scrambler on the transmit datapath
RW
7 bit scrambler seed used when TX scrambler is enabled.
(Default 7’b0000000)
RW
RX_DESCRAMBLER_EN
0 = Disables De-scrambler on the receive datapath (Default 1’b0)
1 = Enables De-scrambler on the receive datapath
RW
RX_DESCRAMBLER_SEED[6:0]
7 bit De-scrambler seed used when RX De-scrambler is enabled.
(Default 7’b0000000)
RW
9.14:8 TX _SCRAMBLER_SEED[6:0]
9.7
Default: 0x0000
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
53
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-15. ARS_CONTROL_1
Address: 0x0A
BIT(s)
NAME
Default: 0x0800
DESCRIPTION
ACCESS
10.15
ARS_RX_CLK_EN
Valid only when ARS function is enabled.
0 = Holds parallel output clock RXCLK_x output fixed at zero until serial data
rate is determined by the ARS. (Default 1’b0)
1 = Allows RXCLK_x output to toggle normally.
RW
10.14
ARS_ LOCK_OVERLAY
Valid only when ARS function is enabled.
0 = LOSx pin does not reflect ARS loss of lock status (Default 1’b0)
1 = Allows ARS loss of lock status to be reflected on LOSx pin
RW
ARS_ EN[1:0]
ARS enable software control. Applicable when RATE_x[2:0] pins are set to
3’b100.
00 = Serdes settings are determined by MDIO (Default 2’b00)
01 = Enable ARS function in respective channel.
Serdes settings for respective channel are determined by ARS in the same
channel. Refer RATE_x[2:0] 3’b101 setting for more information.
RW
10.13:12
10 = Enable ARS function in respective channel.
Serdes settings for respective channel are determined by ARS in the same
channel. Refer RATE_x[2:0] 3’b110 setting for more information.
11 = Enable ARS function in respective channel as slave mode.
If ARS is enabled in partner channel, serdes settings in this channel are
determined by ARS in partner channel.
If ARS is not enabled in partner channel, serdes settings in this channel are
determined by MDIO.
Refer RATE_x[2:0] 3’b111 setting for more information.
54
10.11
ARS_TX_MDIO_GATE
Valid only when ARS_TX_DATAPATH_OVERRIDE (10.10) is set
0 = When ARS successfully determines incoming serial data rate, TX_FIFO
is automatically reset and parallel data is serialized and sent through serial
output pins.
1 = When ARS successfully determines incoming serial data rate, TX FIFO
needs to be manually reset by writing to register bit 4.2 to enable proper
serialization of the parallel data. (Default 1’b1)
RW
10.10
ARS_TX_DATAPATH_OVERRIDE
Applicable during serial data rate determination by ARS
0 = Transmit parallel input data is serialized as received and sent through
serial output pins. (Default 1’b0)
1 = 10 bit data specified in ARS_TX_DATA (10.9:0) is serialized and sent
through serial output pins.
RW
10.9:0
ARS_TX_DATA[9:0]
10 bit data to be serialized during serial rate determination by ARS.
If TX_PMA_BIT_ORDER (8.3) is set, ARS_TX_DATA [9] is serially
transmitted first else ARS_TX_DATA [0] is serially transmitted first.
RW
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-16. ARS_CONTROL_2
Address: 0x0B
BIT(s)
Default: 0x7FFF
NAME
DESCRIPTION
ACCESS
11.15:14
ARS_REF_FREQ[1:0]
Input reference clock frequency selection in ARS mode.
00 = If input reference clock frequency is 122.88 MHz.
01 = If input reference clock frequency is 153.60 MHz. (Default 2’b01)
10 = Reserved
11 = If input reference clock frequency is 307.20 MHz.
RW
11.13:6
ARS_SBR_ENABLE[7:0]
Control to enable rate determination through ARS for each of the 8 supported bit
rates. ARS_SBR_ENABLE[7] for the highest serial bit rate and
ARS_SBR_ENABLE[0] for the lowest serial bit rate. Refer Table 2-6 for supported
serial bit rates.
RW
RESERVED
For TI use only.
RW
ARS_INTERVAL[20:16]
5 MSB’s of 21 bit wide counter defined in terms of number of REFCLK cycles to
determine amount of time that ARS state machine needs to stay in a particular
setting to achieve rate determination.
Wait time is calculated as ARS_INTERVAL[20:0] × 1024 REFCLK periods.
If ARS state machine does not achieve rate determination within the wait time
specified by this counter, ARS state machine will move into next lower serial rate
setting to achieve rate determination.
RW
11.5
11.4:0
Table 3-17. ARS_CONTROL_3
Address: 0x0C
BIT(s)
12.15:0
Default: 0xFFFF
NAME
ARS_INTERVAL[15:0]
DESCRIPTION
ACCESS
16 LSB’s of 21 bit wide counter defined in terms of number of REFCLK cycles to
determine amount of time that ARS state machine needs to stay in a particular
setting to achieve rate determination.
RW
Wait time is calculated as ARS_INTERVAL[20:0] × 1024 REFCLK periods.
If ARS state machine does not achieve rate determination within the wait time
specified by this counter, ARS state machine will move into next lowest serial rate
setting to achieve rate determination.
Table 3-18. ARS_CONTROL_4
Address: 0x0D
BIT(s)
Default: 0x3000
NAME
DESCRIPTION
13.15
RESERVED
For TI use only.
13.14
RESERVED
For TI use only.
13.13
RESERVED
For TI use only.
13.12
RESERVED
For TI use only.
13.11:0
RESERVED
For TI use only.
ACCESS
RW
Table 3-19. ERROR_COUNTER
Address: 0x0E
BIT(s)
14.15:0
Default: 0xFFFD
NAME
ERROR_COUNTER[15:0]
DESCRIPTION
ACCESS
In functional mode if 8b/10b decoder is enabled, this counter reflects number of
invalid code words (includes disparity errors) received by decoder.
COR
In test pattern verification mode (7.12 = 1’b1), this counter reflects error count for the
test pattern selected through 7.10:8
When PRBS_EN pin is set, this counter reflects error count for selected PRBS
pattern. Counter value cleared to 16’h0000 when read.
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
55
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-20. SPI_CONTROL_STATUS
Address: 0x0F
BIT(s)
Default: 0x0205
NAME
DESCRIPTION
ACCESS
15.14
SCL_W
SCL write value
0 = 0 Driven onto SCL line (Default 1’b0)
1 = 1 Driven onto to SCL line
RW
15.13
SCL_GZ
SCL output enable control
0 = SCL treated as output (Default 1’b0)
1 = SCL treated as input. Read value stored in SCL_R (15.12)
RW
15.12
SCL_R
SCL read value (current value of device pin)
0 = 0 Current input value on SCL line
1 = 1 Current input value on SCL line
RO
15.10
SDO_W
SDO write value
0 = 0 Driven onto SDO line (Default 1’b0)
1 = 1 Driven onto to SDO line
RW
15.9
SDO_GZ
SDO output enable control
0 = SDO treated as output
1 = SDO treated as input. Read value stored in SDO_R (15. 8) (Default 1’b1)
RW
15.8
SDO_R
SDO read value (current value of device pin)
0 = 0 Current input value on SDO line
1 = 1 Current input value on SDO line
RO
15.6
SDI_W
SDI write value
0 = 0 Driven onto SDI line (Default 1’b0)
1 = 1 Driven onto to SDI line
RW
15.5
SDI_GZ
SDI output enable control
0 = SDI treated as output (Default 1’b0)
1 = SDI treated as input. Read value stored in SDI_R (15.4)
RW
15.4
SDI_R
SDI read value (current value of device pin)
0 = 0 Current input value on SDI line
1 = 1 Current input value on SDI line
RO
15.2
CS_N_W
CS_N write value
0 = 0 Driven onto CS_N line
1 = 1 Driven onto to CS_N line (Default 1’b1)
RW
15.1
CS_N_GZ
CS_N output enable control
0 = CS_N treated as output (Default 1’b0)
1 = CS_N treated as input. Read value stored in CS_N_R (15.0)
RW
15.0
CS_N_R
CS_N read value (current value of device pin)
0 = 0 Current input value on CS_N line
1 = 1 Current input value on CS_N line
RO
Table 3-21. LATENCY_MEASURE_CONTROL
Address: 0x10
BIT(s)
DESCRIPTION
ACCESS
16.10
RESERVED
For TI use only.
16.9
RESERVED
For TI use only.
16.8
RESERVED
For TI use only.
LATENCY_MEAS_CLK_DIV[1:0]
Latency measurement clock divide control. Valid only when bit 16.2 is 0.
Divides clock to needed resolution. Higher the divide value, lesser the latency
measurement resolution
00 = Divide by 1 (Default 2’b00) (Most Accurate Measurement)
01 = Divide by 2
10 = Divide by 4
11 = Divide by 8 (Longest Measurement Capability)
See Table 2-6
RW
LATENCY_MEAS_CLK_SEL
Latency measurement clock selection.
0 = Selects clock listed in Table 2-6. Bits 16.5:4 can be used to divide this clock
to achieve needed resolution. (Default 1’b0)
1 = Selects respective channel recovered byte clock (Frequency = Serial bit
rate/ 10).
RW
16.5:4
16.2
56
Default: 0x0000
NAME
PROGRAMMERS REFERENCE
RW
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-21. LATENCY_MEASURE_CONTROL (continued)
Address: 0x10
BIT(s)
Default: 0x0000
NAME
DESCRIPTION
ACCESS
16.1
LATENCY_MEAS_EN
Latency measurement enable
0 = Disable Latency measurement (Default ’b0)
1 = Enable Latency measurement
RW
16.0
LATENCY_MEAS_CH_SEL
Latency measurement channel selection
0 = Selects Latency measurement for channel A (Default ’b0)
1 = Selects Latency measurement for channel B
RW
Table 3-22. LATENCY_COUNTER_2
Address: 0x11
BIT(s)
NAME
DESCRIPTION
ACCESS
RO/LH (1)
17.15
LATENCY_MEAS_TX_COMMA
Comma indication for the current latency measurement. Cleared when
Register 0x12 is read
0 = Indicates Latency measurement detected comma on TXD[19:10]
1 = Indicates Latency measurement detected comma on TXD[9:0]
17.14
LATENCY_MEAS_RX_COMMA
Comma indication for the current latency measurement. Cleared when
Register 0x12 is read
0 = Indicates Latency measurement detected comma on RXD[19:10]
1 = Indicates Latency measurement detected comma on RXD[9:0]
17.4
LATENCY_ MEAS_READY
Latency measurement ready indicator
0 = Indicates latency measurement not complete.
1 = Indicates latency measurement is complete and value in latency
measurement counter (LATENCY_MEAS_COUNT[19:0]) is ready to be read.
LATENCY_MEAS_COUNT[19:16]
Bits[19:16] of 20 bit wide latency measurement counter.
Latency measurement counter value represents the latency in number of clock
cycles. This counter will return 20’h00000 if it is read before rx comma is
received. If latency is more than 20’hFFFFF clock cycles then this counter
returns 20’hFFFFF.
17.3:0
(1)
Default: 0x0000
COR (1)
User has to make sure Register 0x11 has to be read first before reading Register 0x12. Latency measurement counter value resets to
20’h00000 when Register 0x12 is read. Comma indication (17.15 and 17.14) and count valid (17.4) bits are also cleared when Register
0x12 is read.
Table 3-23. LATENCY_COUNTER_1
Address: 0x12
Default: 0x0000
BIT(s)
NAME
18.15:0
LATENCY_MEAS_COUNT[15:0]
(1)
DESCRIPTION
ACCESS
Bits[15:0] of 20 bit wide latency measurement counter.
COR (1)
User has to make sure Register 0x11 has to be read first before reading Register 0x12. Latency measurement counter value resets to
20’h00000 when Register 0x12 is read. Comma indication (17.15 and 17.14) and count valid (17.4) bits are also cleared when Register
0x12 is read.
Table 3-24. TI_RESERVED_CONTROL_1
Address: 0x13
BIT(s)
Default: 0x0200
NAME
DESCRIPTION
19.12
RESERVED
For TI use only.
19.9
RESERVED
For TI use only
19.8
RESERVED
For TI use only
19.7
RESERVED
For TI use only
19.3
RESERVED
For TI use only
19.0
RESERVED
For TI use only
ACCESS
RW
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
57
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-25. SERDES_CONTROL_2
Address: 0x14
Default: 0x7C4F
BIT(s)
NAME
DESCRIPTION
20.15
RESERVED
For TI use only.
ACCESS
20.14
RESERVED
For TI use only
20.13
RESERVED
For TI use only
20.12
RESERVED
For TI use only
20.11
RESERVED
For TI use only
20.10
RESERVED
For TI use only
20.9
RESERVED
For TI use only.
20.8
RESERVED
For TI use only.
20.7
RESERVED
For TI use only.
20.6
RESERVED
For TI use only.
20.5:4
RESERVED
For TI use only
20.3
RESERVED
For TI use only
20.2
ENRX
Serdes receiver enable control. Serdes receiver is automatically disabled when PD_TRXx_N
is asserted LOW or when register bit 1.15 is set HIGH.
0 = Disables serdes receiver
1 = Enables serdes receiver (Default 1’b1)
RW
20.1
ENTX
Serdes transmitter enable control. Serdes transmitter is automatically disabled when
PD_TRXx_N is asserted LOW or when register bit 1.15 is set HIGH.
0 = Disables serdes transmitter
1 = Enables serdes transmitter (Default 1’b1)
RW
20.0
ENPLL
Serdes PLL enable control. Serdes PLL is automatically disabled when PD_TRXx_N is
asserted LOW or when register bit 1.15 is set HIGH.
0 = Disables PLL in serdes
1 = Enables PLL in serdes (Default 1’b1)
RW
RW
Table 3-26. TI_RESERVED_CONTROL_3
Address: 0x15
BIT(s)
Default: 0x0023
NAME
DESCRIPTION
21.12
RESERVED
For TI use only.
21.11
RESERVED
For TI use only
21.10
RESERVED
For TI use only
21:9
RESERVED
For TI use only
21.8
RESERVED
For TI use only
21.7
RESERVED
For TI use only
21.6:4
RESERVED
For TI use only
21.2:0
RESERVED
For TI use only
ACCESS
RW
Table 3-27. TI_RESERVED_CONTROL_4
Address: 0x16
58
Default: 0x0023
BIT(s)
NAME
22.12
RESERVED
For TI use only.
DESCRIPTION
22.11
RESERVED
For TI use only
22.10
RESERVED
For TI use only
22.9
RESERVED
For TI use only
22.8
RESERVED
For TI use only
22.7
RESERVED
For TI use only
22.6:4
RESERVED
For TI use only.
22.2:0
RESERVED
For TI use only.
PROGRAMMERS REFERENCE
ACCESS
RW
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-28. TI_RESERVED_CONTROL_5
Address: 0x17
Default: 0x0008
BIT(s)
NAME
DESCRIPTION
23.15
RESERVED
For TI use only.
23.14
RESERVED
For TI use only
23.13:8
RESERVED
For TI use only
23.7:5
RESERVED
For TI use only
23.3
RESERVED
For TI use only
ACCESS
RW
Table 3-29. TI_RESERVED_STATUS_1
Address: 0x18
Default: 0x0000
BIT(s)
NAME
DESCRIPTION
24.13
RESERVED
For TI use only.
24.12
RESERVED
For TI use only
24.9:4
RESERVED
For TI use only
24.3
RESERVED
For TI use only
24.2
RESERVED
For TI use only
24.1
RESERVED
For TI use only
24.0
RESERVED
For TI use only.
ACCESS
RO
Table 3-30. TI_RESERVED_CONTROL_6
Address: 0x19
BIT(s)
Default: 0x0000
NAME
DESCRIPTION
25.6
RESERVED
For TI use only.
25.5:4
RESERVED
For TI use only
25.3:0
RESERVED
For TI use only
ACCESS
RW
Table 3-31. TI_RESERVED_CONTROL_7
Address: 0x1A
BIT(s)
NAME
26.15:0
RESERVED
Default: 0x0000
DESCRIPTION
ACCESS
For TI use only.
RW
Table 3-32. TI_RESERVED_STATUS_2
Address: 0x1B
BIT(s)
NAME
27.15:0
RESERVED
Default: 0x0000
DESCRIPTION
ACCESS
For TI use only.
RO
Table 3-33. TI_RESERVED_CONTROL_8
Address: 0x1C
Default: 0x0001
BIT(s)
NAME
DESCRIPTION
28.14
RESERVED
For TI use only.
28.13
RESERVED
For TI use only
28.12
RESERVED
For TI use only
28.8
RESERVED
For TI use only
28.7:4
RESERVED
For TI use only
28.0
RESERVED
For TI use only
ACCESS
RW
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
59
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 3-34. TI_RESERVED_STATUS_3
Address: 0x1D
BIT(s)
Default: 0x0200
NAME
DESCRIPTION
29.9
RESERVED
For TI use only.
29.8:4
RESERVED
For TI use only
29.3:0
RESERVED
For TI use only
ACCESS
RO
Table 3-35. TI_RESERVED_CONTROL_9
Address: 0x1E
BIT(s)
NAME
30.15:0
RESERVED
Default: 0x0000
DESCRIPTION
ACCESS
For TI use only.
RW
Table 3-36. TI_RESERVED_CONTROL_10
Address: 0x1F
BIT(s)
NAME
31.15:0
RESERVED
Default: 0x0000
DESCRIPTION
ACCESS
For TI use only.
RW
Table 3-37. TI_RESERVED_STATUS_4
Address: 0x8000
BIT(s)
Default: 0x0000
NAME
32768.15:0
RESERVED
DESCRIPTION
ACCESS
For TI use only.
RO
Table 3-38. TI_RESERVED_STATUS_5
Address: 0x8001
BIT(s)
Default: 0x0000
NAME
32769.15:0
RESERVED
DESCRIPTION
ACCESS
For TI use only.
RO
Table 3-39. TI_RESERVED_STATUS_6
Address: 0x8002
BIT(s)
Default: 0x0000
NAME
32770.15:0
RESERVED
DESCRIPTION
ACCESS
For TI use only.
RO
Table 3-40. TI_RESERVED_STATUS_7
Address: 0x8003
BIT(s)
Default: 0x0000
NAME
32771.15:0
RESERVED
DESCRIPTION
ACCESS
For TI use only.
RO
Table 3-41. TI_RESERVED_STATUS_8
Address: 0x8004
BIT(s)
Default: 0x0000
NAME
32772.15:0
RESERVED
DESCRIPTION
ACCESS
For TI use only.
RO
Table 3-42. TI_RESERVED_STATUS_9
Address: 0x8005
BIT(s)
32773.15:0
60
Default: 0x0000
NAME
RESERVED
DESCRIPTION
For TI use only.
PROGRAMMERS REFERENCE
ACCESS
RO
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 3-43. TI_RESERVED_STATUS_10
Address: 0x8006
BIT(s)
Default: 0x0000
NAME
32774.15:0
RESERVED
DESCRIPTION
ACCESS
For TI use only.
RO
Table 3-44. TI_RESERVED_STATUS_11
Address: 0x8007
BIT(s)
32775.15:0
3.1
Default: 0x0000
NAME
RESERVED
DESCRIPTION
ACCESS
For TI use only.
RO
LL = Latched Low
Latched low means that if a condition is occurring, the register bit will read low. Latched low also means
that if a condition has occurred since the last time the register was read, it will read low. If a latched low
register bit reads high, it means that the condition is not occurring presently, and it has not occurred since
the last time the register was read. A latched low register, when read low, should be read again to
distinguish if a condition occurred previously or is still occurring. If it occurred previously, the second read
will read high. If it is still occurring, the second read will read low.
3.2
LH = Latched High
Latched high means that if a condition is occurring, the register bit will read high. Latched high also means
that if a condition has occurred since the last time the register was read, it will read high. If a latched high
register bit reads low, it means that the condition is not occurring presently, and it has not occurred since
the last time the register was read. A latched high register, when read high, should be read again to
distinguish if a condition occurred previously or is still occurring. If it occurred previously, the second read
will read low. If it is still occurring, the second read will read high.
3.3
COR = Clear On Read
Counters indicated as COR are cleared after being read.
PROGRAMMERS REFERENCE
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
61
TLK6002
SLLSE34 – MAY 2010
www.ti.com
4 ELECTRICAL SPECIFICATIONS
4.1
ABSOLUTE MAXIMUM RATINGS
over operating free-air temperature range (unless otherwise noted) (1)
(2)
VALUE / UNIT
Supply voltage
Input Voltage, VI
DVDD, AVDD, VDDT, VPP, VDDD
–0.3 to 1.4 V
VDDRA/B, VDDO1/2/3, VDDQA/B, VREFTA/B
–0.3 to 2.2 V
(LVCMOS/LVPECL/HSTL/CML/Analog)
–0.3 to Supply + 0.3 V
Storage temperature
Electrostatic
Discharge
–65°C to 150°C
HBM
1 KV
CDM
500 V
Characterized free-air operating temperature range
–40°C to 85°C
(1)
(2)
Stresses beyond those listed under "absolute maximum ratings" may cause permanent damage to the device. These are stress ratings
only, and functional operation of the device at these or any other conditions beyond those indicated under "recommended operating
conditions" is not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability.
All voltages are with respect to ground (AGND/DGND).
4.2
RECOMMENDED OPERATING CONDITIONS
over operating free-air temperature range (unless otherwise noted)
PARAMETER
VDDD
AVDD
DVDD
VDDT
VPP
Digital / Analog Supply Voltages
VREFTA
VREFTB
HSTL Voltage Reference Accuracy
VDDQA/B
HSTL I/O Supply Voltages
VDDRA
VDDRB
SERDES PLL Regulator Voltage
VDDO1/2/3
LVCMOS I/O Supply Voltage
IDD
Supply
Current
CONDITION
VDDQA/2 or VDDQB/2 Voltage
Variance
UNIT
0.95
1.05
V
1.00
–1%
1%
1.4
1.5
1.6
1.8V Nominal
1.7
1.8
1.9
1.5V Nominal
1.425
1.5
1.575
1.8V Nominal
1.71
1.8
1.89
1.5V Nominal
1.425
1.5
1.575
1.8V Nominal
1.71
1.8
1.89
VDDD
75
AVDD
250
DVDD + VPP
200
VDDT
6.144 Gbps
VDDQA/B (1.5V /1.8V Mode)
650/800
V
V
mA
5/10
VREFTA + VREFTB (1.5V /1.8V Mode)
All Supplies Worst Case
15
V
40/40
VDDO1+VDDO2+VDDO3 (1.5V /1.8V Mode)
62
MAX
1.5V Nominal
VDDRA + VDDRB (1.5V /1.8V Mode)
PD
MIN NOM
0.5/0.5
See Table 4-3.
ELECTRICAL SPECIFICATIONS
W
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
RECOMMENDED OPERATING CONDITIONS (continued)
over operating free-air temperature range (unless otherwise noted)
PARAMETER
CONDITION
MIN NOM
MAX
VDDD
28
AVDD
15
DVDD + VPP
62
VDDT
ISD
Shutdown
Current
1
VDDQA/B (1.5V Mode /1.8V Mode)
5/5
PD* Asserted
VDDRA + VDDRB (1.5V Mode /1.8V Mode)
1/1
VDDO1+VDDO2+VDDO3 (1.5V Mode /1.8V
Mode)
1/2
Reference Clock Timing Requirement (REFCLK_0/1_P/N)
PARAMETER
CONDITION
MIN NOM MAX
Frequency
122.88
Accuracy
100
Relative to Incoming Serial Data Rate
–200
200
Synchronous (Multiple/Divide)
High Time
Jitter
Random and deterministic
CIN
Input Capacitance
RIN
Input Differential Impedance
Trise
Rise Time
CONDITION
0
55%
MHz
ppm
ppm
ps
MIN NOM
MAX
UNIT
250
2000
mVdfpp
3
80
20% to 80%
100
50
pF
120
Ω
600
ps
Differential Clock Output Electrical Characteristics (CLK_OUT_P/N)
PARAMETER
CONDITION
MIN NOM
Vod
Differential Output Voltage
Peak to peak
TRise
Output Rise Time
10 to 90% 2pF lumped C loadAC Coupled
RTERM
Output Termination
CLK_OUT_P/N to DVDD
FMAX
Output Frequency
1000
40
0
50
MAX
UNIT
2000
mVdfpp
180
ps
60
Ω
625
MHz
MAX
UNIT
LVCMOS Electrical Characteristics (VDDO = VDDO1/VDDO2/VDDO3)
PARAMETER
VOL
0
50%
40
PARAMETER
Differential Input Voltage
VOH
0
45%
UNIT
Differential Reference Clock Electrical Characteristics (REFCLK_0/1_P/N)
Vid
4.6
800
–100
Duty Cycle
4.5
–
Relative to Nominal Serial Data Rate
Accuracy to TXCLK_A/B
4.4
mA
0.05/0.0
5
VREFTA + VREFTB (1.5V /1.8V Mode)
4.3
UNIT
High-level output voltage
Low-level output voltage
CONDITION
MIN
NOM
IOH = 2 mA, Driver Enabled (1.8 V)
VDDO – 0.45
VDDO
IOH = 2 mA, Driver Enabled (1.5 V)
0.75×VDDO
VDDO
IOL = –2 mA, Driver Enabled (1.8 V)
0
0.45
IOL = –2 mA, Driver Enabled (1.5 V)
0
0.25×VDDO
V
V
VIH
High-level input voltage
0.65×VDDO
VDDO+0.3
VIL
Low-level input voltage
-0.3
0.35×VDDO
V
IIH, IIL
Receiver Only
±170
µA
Low/High Input Current
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
V
63
TLK6002
SLLSE34 – MAY 2010
www.ti.com
LVCMOS Electrical Characteristics (VDDO = VDDO1/VDDO2/VDDO3) (continued)
PARAMETER
CONDITION
Driver Only
Driver Disabled
IOZ
Driver/Receiver With Pullup/Pulldown
Driver DisabledWith Pull Up/Down
Enabled
CIN
Input capacitance
4.7
MIN
NOM
MAX
UNIT
±25
µA
±195
µA
3
pF
HSTL Signals Electrical Characteristics (VDDQA/B = 1.5/1.8V)
PARAMETER
CONDITION
MIN
NOM
UNIT
High-level output voltage
VOL(dc)
Low-level output voltage
VOH(ac)
High-level output voltage
VOL(ac)
Low-level output voltage
VIH(dc)
High-level DC input voltage
DC input, logic high
VIL(dc)
Low-level DC input voltage
DC input, logic low
VIH(ac)
High-level AC input voltage
AC input, logic high
VIL(ac)
Low-level AC input voltage
AC input, logic low
IOH(dc)
High output current
8
IOL(dc)
Low output current
–8
IIH
Input High Current
10
µA
IIL
Input Low Current
–10
µA
CIN
Input Capacitance
4
pF
Tacr
AC Test Condition
Rise Time Rate (In 20 → 80% Swing
Region)
1
1
1
ns/V
Tacs
AC Test Condition
Signal Swing
1
1
1
V
4.8
VDDQA/B – 0.4
MAX
VOH(dc)
V
0.40
V
VDDQA/B + 0.3
V
0.50
V
VREFTA/B+ 0.10
VDDQA/B + 0.3
V
-0.30
VREFTA/B – 0.1
V
VREFTA/B + 0.20
VDDQA/B + 0.3
V
–0.30
VREFTA/B –
0.20
V
VDDQA/B – 0.5
TX Output Differential Peak-to-Peak voltage
swing
See Figure 4-1 and Figure 4-2
CONDITION
mA
MIN
NOM
MAX
SWING (2.3:0) = 0000
84
126
160
SWING (2.3:0) = 0001
163
215
253
SWING (2.3:0) = 0010
247
303
347
SWING (2.3:0) = 0011
331
395
444
SWING (2.3:0) = 0100
415
478
538
SWING (2.3:0) = 0101
499
572
633
SWING (2.3:0) = 0110
582
662
735
SWING (2.3:0) = 0111
668
756
835
SWING (2.3:0) = 1000
740
839
927
SWING (2.3:0) = 1001
831
932
1020
SWING (2.3:0) = 1010
904
1020
1130
SWING (2.3:0) = 1011
988
1110
1220
SWING (2.3:0) = 1100
1060
1190
1340
SWING (2.3:0) = 1101
1150
1280
1430
SWING (2.3:0) = 1110
1220
1360
1530
SWING (2.3:0) = 1111
1300
1450
1630
Vpre/post
TX Output Pre/Post Cursor Emphasis
Voltage
See register bits TWPOST1 2.12:8 and
TWPRE 2.7:4 for de-emphasis settings.
See Figure 4-2.
VCMT
TX output common mode voltage
See Figure 4-1
64
mA
Serial Transmitter Characteristics
PARAMETER
VOD(pp)
VDDQA/B + 0.3
ELECTRICAL SPECIFICATIONS
–17.5/
–37.5%
UNIT
mVdfpp
+17.5/
+37.5%
VDDT –
[0.25×VOD(pp)]
mV
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Serial Transmitter Characteristics (continued)
PARAMETER
tskew
CONDITION
Output Skew
MIN
NOM
MAX
CPRI LV
15
CPRI HV
25
CPRI LV-II (DCD ≤ 0.05UI)
15
CPRI LV (Note: MIN is not per CPRI)
54
–
–
CPRI HV (Note: MIN is not per CPRI)
55
–
327
CPRI LV-II
30
–
tr, tf
Differential output signal rise, fall time (20%
to 80%)Differential Load = 100Ω
JT
Serial Output Total Jitter
(CPRI LV/LV-II and OBSAI rates)
Serial Rate ≤ 3.072 Gbps(Not Applicable to LV-II)
0.35
Serial Rate > 3.072 Gbps(And All LV-II Rates)
0.30
JD
Serial Output Deterministic Jitter
(CPRI LV/LV-II and OBSAI rates)
Serial Rate ≤ 3.072 Gbps(Not Applicable to LV-II)
0.17
Serial Rate > 3.072 Gbps(And All LV-II Rates)
0.15
JT
Serial Output Total JitterJitter (CPRI
E.6/12.HV)
JD
Serial Output Deterministic Jitter (CPRI
E.6/12.HV)
T(LATENCY)
Transmit Latency
UNIT
ps
ps
–
UI
UI
0.279
CPRI E.6/12.HV(0.6144 and 1.2288 Gbps)
0.14
See latency specifications in Appendix D.
–
0.5* VDE*
VOD(pp)
VCMT
–
–
Bit
Times
0.5*
VOD(pp)
0.25* VDE*VOD(pp)
,
tr tf
bit
time
0.25* VO(pp)
Figure 4-1. Transmit Output Waveform Parameter Definitions
+V0/0
+Vpst
+Vpre
+Vss
0
-Vss
-Vpre
-Vpst
-V0/0
UI
h-1 = TWPRE (0%
-17.5% for typical application) setting
h1 = TWPOST1 (0%
-37.5% for typical application) setting
h0 = 1 - |h1| - |h-1|
V0/0 = Output Amplitude with TWPRE = 0%, TWPOST = 0%.
Vss = Steady State Output Voltage = V0/0 * | h1 + h0 + h-1|
Vpre = PreCursor Output Voltage = V0/0 * | -h1 – h0 + h-1|
Vpst = PostCursor Output Voltage = V0/0 * | -h1 + h0 + h-1|
Figure 4-2. Pre/Post Cursor Swing Definitions
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
65
TLK6002
SLLSE34 – MAY 2010
www.ti.com
+ 800 mV
+ A2
+ 400 mV
+ A1
0 V
0
- 400 mV
- A1
- 800 mV
- A2
0
0 .175
0
X 1
0 .39
0 .61
0 .825
1
X2
1-X 2
1 -X 1
1
Figure 4-3. CPRI TX LV Mask for: E.6/12/24/30.LV (0.6144/1.2288/2.4576/3.072 Gbps)
+ 1000 mV
+ A2
+ 550 mV
+ A1
0 V
0
- 550 mV
- A1
- 1000 mV
- A2
0
0 .14
0
X1
0 .34
0 .66
0 .86
1
X2
1-X 2
1 - X1
1
Figure 4-4. CPRI TX HV Output Mask for: E.6/12.HV (0.6144/1.2288 Gbps)
+ 800 mV
+ A2
+ 200 mV
+ A1
0 V
0
- 200 mV
- A1
- 800 mV
- A2
0
0 .175
0
X 1
0 .39
0 .61
0 .825
1
X2
1-X 2
1 -X 1
1
Figure 4-5. OBSAI TX Output Mask for Rates ≤ 3.072 Gbps
66
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
+ 600 mV
+ A2
+ 400 mV
+ A1
0 V
0
- 400 mV
- A1
- 600 mV
- A2
0
0 .15
0 .4
0 .6
0 .85
0
X 1
X2
1-X 2
1 -X 1
1
1
Figure 4-6. OBSAI 6.144 Gbps and CPRI (LV-II) (All Rates) TX Output Mask
Note: Due to process variation, the output mask cannot always be achieved with a common setting for all
devices. However, for a given device, a particular swing setting will pass output mask requirements.
4.9
Serial Receiver Characteristics
PARAMETER
CONDITION
VID
RX input differential voltage|RXP – RXN|
VID(pp)
RX input differential peak-to-peak voltage
swing
2×|RXP – RXN|
CI
RX input capacitance
JTOL
JDR
T(LATENCY)
MIN
NOM
MAX
UNIT
mV
Full Rate AC Coupled
50
600
Half/Quarter/Eighth Rate AC Coupled
50
800
Full Rate AC Coupled
100
1200
Half/Quarter/Eighth Rate AC Coupled
100
1600
2
Jitter Tolerance, Total Jitter at Serial Input
(DJ + RJ)(BER 10-15)
Serial Input Deterministic Jitter(BER 10-15)
Zero crossing Half/Quarter/Eighth Rate
0.66
Zero crossing Full Rate
0.65
Zero crossing Half/Quarter/Eighth Rate
0.50
Zero crossing Full Rate
pF
UIp-p
0.35
See latency specifications in Appendix D
Receive Latency
mVdfpp
–
–
–
Bit
Times
+ A2
+ 800 mV
+ A1
+ 100 mV
0V
0
- 100 mV
- A1
- 800 mV
- A2
0
0
0 .275
X
0 .4
X 2
0.6
0 .725
1-X2 1-X1
1
1
Figure 4-7. CPRI LV Receiver Mask for E.6/12/24/30.LV (0.6144/1.2288/2.4576/3.072 Gbps)
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
67
TLK6002
SLLSE34 – MAY 2010
www.ti.com
+ 1000 mV
+ A2
+ 200 mV
+ A1
0V
0
- A1
- 200 mV
- 1000 mV
- A2
0
0 .33
0
X
0 .5
X 2
0.5
0 .67
1
1-X2 1-X1
1
Figure 4-8. CPRI HV Receiver Mask for E.6/12.HV (0.6144/1.2288 Gbps)
+ A2
+ 800 mV
+ A1
+ 100 mV
0
0 V
- 100 mV
- A1
- 800 mV
- A2
0
0 .275
0 .5
0 .5 0 .725
1
0
X1
X2
1-X2 1-X1
1
Figure 4-9. OBSAI Receiver Mask for Rates ≤ 3.072 Gbps
+ A2
+ 600 mV
+ 50 mV
+ A1
0V
0
- 50 mV
- A1
- 600 mV
- A2
0
0 .3
0 .5
0 .5
0 .7
0
X1
X2
1- X2 1- X1
1
1
Figure 4-10. OBSAI 6.144 Gbps and CPRI LV-II (All Rates) Receiver Mask
68
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Jitter Tolerance:
The peak to peak total jitter tolerance for the RP3 receiver is 0.65 UI. This total jitter is composed of three
components; deterministic jitter, random jitter, and an additional sinusoidal jitter.
The deterministic jitter tolerance is 0.37 UI minimum. The sum of deterministic and random jitter is 0.55 UI
minimum. The additional sinusoidal jitter which the receiver must tolerate will have frequencies and amplitudes
conforming to the mask presented in Figure 4-11 and Table 4-1.
UI 2pp
Sinusoidal
Jitter
Amplitude
(UI )
UI1pp
f1
Frequency
f2
20 MHz
Figure 4-11. OBSAI Sinusoidal Jitter Mask
Table 4-1. Sinusoidal Jitter Mask Values
Frequency
(MBaud)
f1
(kHz)
f2
(kHz)
UI1pp
UI2pp
768
5.4
460.8
0.1
8.5
1536
10.9
921.6
0.1
8.5
3072
21.8
1843.2
0.1
8.5
6114
36.9
3686
0.05
8.5
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
69
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Input Jitter Definition
JDR
JR
JR
JTOL
NOTE: JTOL = JR + JDR, where JTOL is the receive jitter tolerance, JDR is the received deterministic jitter, and JR is the
Gaussian random edge jitter distribution at a maximum BER = 10-12 for CPRI link and BER = 10-15 for OBSAI (RP3)
link.
4.10
HSTL Output Switching Characteristics (DDR Timing Mode Only)
(over recommended operating conditions unless otherwise noted).
PARAMETER
CONDITION
MIN
MAX
UNIT
tsetup
RDx (1) setup prior to RXCLK_x transition high or
low
Source Centered. See Figure 4-12 (2)
0.15 ×
tperiod
ps
thold
RDx hold after RXCLK_x transition high or low
Source Centered. See Figure 4-12 (2)
0.15 ×
tperiod
ps
Tduty
RXCLK_x (3) Duty Cycle
Source Centered and Source Aligned. (2)
tperiod
RXCLK_x Period
Source Centered and Source Aligned.
6.4
85.11
ns
Tfreq
RXCLK_x Frequency
Source Centered and Source Aligned.
11.75
156.25
MHz
Tpd
RXCLK_x rising or falling to RDx valid.
Source Aligned See Figure 4-13 (2)
(1)
(2)
(3)
45%
55%
–0.10 × +0.10 ×
tperiod tperiod
ps
RDx refers to either RDA or RDB for channels A and B respectively
Cload = 10pF, using timing reference of (VDDQA/B)/2.
RXCLK_x refers to RXCLK_A or RXCLK_B for channels A and B respectively.
tPERIOD
VOH(ac)
RXCLK_x
VDDQ/2
VOL(ac)
tSETUP
tHOLD
tSETUP
tHOLD
VOH(ac)
RDx
VDDQ/2
VOL(ac)
Figure 4-12. HSTL (DDR Timing Mode Only) Source Centered Output Timing Requirements
70
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
VOH(ac)
RXCLK_x
VDDQ/2
VOL(ac)
Tpd
Tpd
VOH(ac)
VDDQ/2
RDx
VOL(ac)
Figure 4-13. HSTL (DDR Timing Mode Only) Source Aligned Output Timing Requirements
4.11
HSTL Output Switching Characteristics (SDR Timing Mode Only)
PARAMETER
CONDITION
MIN
MAX
40%
60%
3.2
42.55
ns
23.5
312.5
MHz
–0.10 ×
tperiod
+0.10 ×
tperiod
ps
+0.10 ×
tperiod
ps
Tduty
RXCLK_x (1) Duty Cycle
Rising and Falling Edge Aligned Data (2)
tperiod
RXCLK_x Period
Rising and Falling Edge Aligned Data
Tfreq
RXCLK_x Frequency
Rising and Falling Edge Aligned Data
RXCLK_x rising to RDx (3) valid.
Rising Edge Aligned, See Figure 4-14. (2)
RXCLK_x falling to RDx valid.
Falling Edge Aligned, See Figure 4-15. (2)
–0.10 ×
tperiod
Tpd
(1)
(2)
(3)
UNIT
RXCLK_x refers to RXCLK_A or RXCLK_B for channels A and B respectively.
Cload = 10pF, using timing reference of (VDDQA/B)/2.
RDx refers to either RDA or RDB for channels A and B respectively.
tPERIOD
VOH(ac)
VDDQ/2
VOL(ac)
RXCLK_x
TPD
VOH(ac)
RDx VDDQ/2
VOL(ac)
Figure 4-14. HSTL (SDR Timing Mode Only) Rising Edge Aligned Output Timing Requirements
tPERIOD
VOH(ac)
VDDQ/2
RXCLK_x
VOL(ac)
TPD
VOH(ac)
RDx VDDQ/2
VOL(ac)
Figure 4-15. HSTL (SDR Timing Mode Only) Falling Edge Aligned Output Timing Requirements
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
71
TLK6002
SLLSE34 – MAY 2010
4.12
www.ti.com
HSTL (DDR Timing Mode Only) Input Timing Requirements
PARAMETER
(1)
CONDITION
(2)
tsetup
TDx setup prior to TXCLK_x
or low
thold
TDx hold after TXCLK_x transition high or low
tduty
transition high
TXCLK_x Duty Cycle
MIN
MAX
UNIT
Source Centered, See Figure 4-16
and (3)
0.075 × tperiod
ps
Source Centered, See Figure 4-16
and (3)
0.075 × tperiod
ps
Source Centered, See Figure 4-16
and (3)
40%
60%
Source Aligned, See Figure 4-17
and (3)
45%
55%
tperiod
TXCLK_x Period
Source Centered and Aligned
6.4
85.11
ns
Tfreq
TXCLK_x Frequency
Source Centered and Aligned
11.75
156.25
MHz
Tskew (4) (5
TXCLK_x rising or falling to TDx valid.
Source Aligned, See Figure 4-17
and (3)
–0.14 × tperiod
+0.14 × tperiod
)
(1)
(2)
(3)
(4)
(5)
ps
TDx refers to either channel A (TDA) or B (TDB).
TXCLK_x refers to either channel A (TXCLK_A) or channel B (TXCLK_B).
Input timing reference of (VDDQA/B)/2 with ±1 ns/V rise time on all input signals.
When Tfreq ≤60 MHz, Tskew Minimum is –3ns.
When Tfreq ≤60 MHz, Tskew Maximum is 3ns.
tPERIOD
VIH(ac)
TXCLK_x
VDDQ/2
VIL(ac)
tSETUP
tHOLD
tSETUP
tHOLD
VIH(ac)
TDx
VDDQ/2
VIL(ac)
Figure 4-16. HSTL (DDR Timing Mode Only) Source Centered Data Input Timing Requirements
VOH(ac)
TXCLK_x
VDDQ/2
VOL(ac)
TSKEW
TSKEW
VOH(ac)
TDx VDDQ/2
VOL(ac)
Figure 4-17. HSTL (DDR Timing Mode Only) Source Aligned Data Input Timing Requirements
72
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
4.13 HSTL (SDR Timing Mode Only) Input Timing Requirements
PARAMETER
(1)
transition
MIN MAX
tsetupH
TDx
high
tholdH
TDx hold after TXCLK_x transition high
tsetupL
TDx setup prior to TXCLK_x transition low
tholdL
TDx hold after TXCLK_x transition low
Rising Edge Aligned (Falling Edge Sampled) Data,
Figure 4-19
tduty
TXCLK_x Duty Cycle
Rising and Falling Edge Sampled Data
tperiod
TXCLK_x Period
Tfreq
TXCLK_x Frequency
(1)
(2)
(3)
setup prior to TXCLK_x
CONDITION
(2)
Falling Edge Aligned (Rising Edge Sampled) Data, (3) See
Figure 4-18
(3)
See
UNIT
480
ps
480
ps
480
ps
480
ps
40%
60%
Rising and Falling Edge Aligned Data
3.2
42.5
5
ns
Rising and Falling Edge Aligned Data
23.5
312.
5
MHz
TDx refers to either channel A (TDA) or B (TDB).
TXCLK_x refers to either channel A (TXCLK_A) or channel B (TXCLK_B).
Input timing reference of (VDDQA/B)/2 with ±1 ns/V rise time on all input signals.
tPERIOD
VIH(ac)
TXCLK_x
VDDQ/2
VIL(ac)
tSETUPH
tHOLDH
VIH(ac)
TDx VDDQ/2
VIL(ac)
Figure 4-18. HSTL (SDR Timing Mode Only) Falling Edge Aligned (Rising Edge Sampled) Data Input
Timing Requirements
tPERIOD
VIH(ac)
VDDQ/2
TXCLK_x
VIL(ac)
tSETUPL
tHOLDL
VIH(ac)
TDx
VDDQ/2
VIL(ac)
Figure 4-19. HSTL (SDR Timing Mode Only) Rising Edge Aligned (Falling Edge Sampled) Data Input
Timing Requirements
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
73
TLK6002
SLLSE34 – MAY 2010
www.ti.com
4.14 MDIO Timing Requirements Over Recommended Operating Conditions (unless
otherwise noted)
PARAMETER
CONDITION
See Figure 4-20
MIN NOM MAX
tperiod
MDC period
tsetup
MDIO setup to ↑ MDC
10
thold
MDIO hold to ↑ MDC
10
Tvalid
MDIO valid from MDC ↑
UNIT
100
0
40
ns
MDC
tPERIOD
tSETUP
tHOLD
MDIO
Figure 4-20. MDIO Read/Write Timing
4.15 JTAG Timing Requirements Over Recommended Operating Conditions (unless
otherwise noted)
PARAMETER
CONDITION
Tperiod
TCK period
Tsetup
TDI/TMS/TRST_N setup to ↑ TCK
Thold
TDI/TMS/TRST_N hold from ↑ TCK
Tvalid
TDO delay from TCK Falling
See Figure 4-21
MIN
NOM
MAX
UNIT
66.67
ns
3
ns
5
0
ns
10
ns
TCK
tPERIOD
tSETUP
tHOLD
TDI /TMS/
TRST_N
tVALID
TDO
Figure 4-21. JTAG Timing
74
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
4.16 Power Sequencing Guidelines
The TLK6002 allows either the core or I/O power supply to be powered up for an indefinite period of time
while the other supply is not powered up, if all of the following conditions are met:
1. All maximum ratings and recommending operating conditions are followed
2. Bus contention while 1.5/1.8V power is applied (>0V) must be limited to 100 hours over the projected
lifetime of the device.
3. Junction temperature is less than 105°C during device operation. Note: Voltage stress up to the
absolute maximum voltage values for up to 100 hours of lifetime operation at a TJ of 105°C or lower
will minimally impact reliability.
The TLK6002 inputs are not failsafe (i.e. cannot be driven with the I/O power disabled). TLK6002 inputs
should not be driven high until their associated power supply is active.
4.17 HSTL Interface
The HSTL interface allows for either 1.5V or 1.8V operation. Source series (output) and parallel end
(input) resistance is dynamically updated to compensate for process, voltage, and temperature. RES*
device pins are referenced to accurately set the impedances.
The source series (HSTL output) impedance is dynamically calibrated to 50 Ω using an external 50 Ω
resistor.
There are three options on parallel end (HSTL input) termination:
1. No end termination. This yields the lowest power dissipation, at the cost of signal integrity
performance.
2. Half Strength Mode – 100 Ω Thevenin Equivalent – This mode is comprised of two 200 Ω resistors,
placed between the input signal and VDDQA/B, and the input signal and DGND. This selection yields a
blend between signal integrity performance and power dissipation.
3. Full Strength Mode – 50 Ω Thevenin Equivalent – This mode is comprised of two 100 Ω resistors,
placed between the input signal and VDDQA/B, and the input signal and DGND. This selection yields
the best signal integrity performance at the cost of highest power dissipation.
All three HSTL input modes can be selected through the MDIO interface on a per channel basis through
register bits (6.1:0). The HSTL output driver slew rate is also selectable, and is selected in register bit
(6.2).
Figure 4-22 shows the schematic of the internal HSTL driver and internal HSTL receiver.
VDDQ
50 W
100 W/200 W/open (W)
50 W transmission line
100 W/200 W/open (W)
VDDQ
GND
1 kW
1 kW
VREF* (VDDQ/2)
GND
OUTPUT
PCB
INPUT
Figure 4-22. HSTL I/O Schematic
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
75
TLK6002
SLLSE34 – MAY 2010
www.ti.com
4.18 Device Initialization
The following sequence should be performed to initialize and ensure proper operation of the TLK6002
device.
4.18.1 20-Bit Interface Mode (8b/10b Encoder/Decoder Disabled) (All CPRI/OBSAI Rates)
Note: Assume both channel A and channel B have the same setup.
REFCLK frequency = 122.88 MHz, Mode = Transceiver, Parallel Interface = 20-Bit SDR Falling Edge
Aligned Mode, RXCLK_A/B out = RXBCLK_A/B, Serial Data Rate is CPRI/OBSAI standard rate as shown
below.
• Device Pin Setting(s) – Pin settings allow for maximum software configurability.
– Ensure CODEA_EN input pin is Low.
– Ensure CODEB_EN input pin is Low.
– Ensure RATE_A[2:0] input pins are 3’b100 (High, Low, Low) to enable software control.
– Ensure RATE_B[2:0] input pins are 3’b100 (High, Low, Low) to enable software control.
– Ensure PD_TRXA_N input pin is High.
– Ensure PD_TRXB_N input pin is High.
– Ensure PRBS_EN input pin is Low.
– Ensure REFCLK_A_SEL input pin is Low to enable software control.
– Ensure REFCLK_B_SEL input pin is Low to enable software control.
• Reset Device
– Issue a hard or soft reset (RESET_N asserted for at least 10 µs -or- Write 1’b1 to 0.15
GLOBAL_RESET) after power supply stabilization.
• Enable MDIO global write so that each MDIO write affects both channels to shorten provisioning time
– Write 1’b1 to 0.11 GLOBAL_WRITE
• Clock Configuration
– Select Channel A SERDES REFCLK input (Default = REFCLK_0_P/N)
• If REFCLK_0_P/N used – Write 1’b0 to 0.1 REFCLK_A_SEL
• If REFCLK_1_P/N used – Write 1’b1 to 0.1 REFCLK_A_SEL
– Select Channel B SERDES REFCLK input (Default = REFCLK_0_P/N)
• If REFCLK_0_P/N used – Write 1’b0 to 0.0 REFCLK_B_SEL
• If REFCLK_1_P/N used – Write 1’b1 to 0.0 REFCLK_B_SEL
• Data Rate Setting (select one of the following 8 cases)
– If serial data rate is 6144.00Mbps: write 2’b00 to 1.7:6 RATE_TX[1:0], write 2’b00 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select FULL rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x010E).
– If serial data rate is 4915.20Mbps: Write 2’b00 to 1.7:6 RATE_TX[1:0], write 2’b00 to 1.5:4
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select FULL rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x010D).
– If serial data rate is 3072.00Mbps: Write 2’b01 to 1.7:6 RATE_TX[1:0], write 2’b01 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select HALF rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x015E).
– If serial data rate is 2457.60Mbps: Write 2’b01 to 1.7:6 RATE_TX[1:0], write 2’b01 to 1.5:4
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select HALF rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x015D).
– If serial data rate is 1536.00Mbps: Write 2’b10 to 1.7:6 RATE_TX[1:0], write 2’b10 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select QUARTER rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x01AE).
– If serial data rate is 1228.80Mbps: Write 2’b10 to 1.7:6 RATE_TX[1:0], write 2’b10 to 1.5:4
76
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
•
•
•
•
•
•
•
•
•
•
SLLSE34 – MAY 2010
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select QUARTER rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x01AD).
– If serial data rate is 768.00Mbps: Write 2’b11 to 1.7:6 RATE_TX[1:0], write 2’b11 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select EIGHTH rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x01FE).
– If serial data rate is 614.40Mbps: Write 2’b11 to 1.7:6 RATE_TX[1:0], write 2’b11 to 1.5:4
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select EIGHTH rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x01FD).
Serial Configuration
– Configure the following bits per the desired application
• 1.9:8 (LOOP_BANDWIDTH[1:0])
• 2.12:8 (TWPOST1[4:0])
• 2.7:4 (TWPRE[3:0])
• 2.3:0 (SWING[3:0])
• 8.14:12 (EQPRE[2:0])
• 8.11:10 (CDRTHR[1:0]) = 2’b01
• 8.9:8 (CDRFMULT[1:0]) = 2’b00
Mode Control
– Channel synchronization (comma enable) is on by default and the parallel output 10 bit codes are
byte aligned.
– The default MDIO register settings should enable TBI SDR Falling Edge Aligned mode. To use a
different parallel IO align mode:
• If SDR Rising Edge Aligned: write 0x0183 to CHANNEL_CONTROL_3 register.
• If DDR Source Centered: write 0x01C0 to CHANNEL_CONTROL_3 register.
• If DDR Source Aligned: write 0x01C3 to CHANNEL_CONTROL_3 register.
Enable desired status signals to LOSA and LOSB for real time monitoring per channel. Any number of
signals can be enabled at once.
– If SERDES Rx Loss of Signal condition monitored: write 1’b1 to 6.10 LOS_OVERLAY.
– If channel synchronization status monitored: write 1’b1 to 6.9 CH_SYNC_OVERLAY.
– If PLL lock status monitored: write 1’b1 to 6.8 PLL_LOCK_OVERLAY.
– If SERDES AGC unlock status monitored: write 1’b1 to 7.7 AGCLOCK_OVERLAY.
– If SERDES AZDONE status monitored: write 1’b1 to 7.6 AZDONE_OVERLAY.
Check SERDES PLL Status for Locked State
– Poll 5.0 PLL_LOCK (per channel) until it is asserted (high)
Toggle ENRX
– Write 1’b0 to 20.2 (ENRX)
– Write 1’b1 to 20.2 (ENRX)
Final CDR Configuration
– Wait until either AGC_LOCKED asserted or 250M UI
– Write 8.9:8 (CDRFMULT[1:0]) = 2’b01
– Poll 5.13 AGC_LOCKED (1’b1) (per channel)
Issue Data path Reset
– Write 1’b1 to 4.3 DATAPATH_RESET
Clear Latched Registers
– Read 5 CHANNEL_STATUS_1 to clear (per channel)
Device provisioning has completed at this point
Periodically Check Device Operational Mode Status (Non-Errored Read Values Shown Below):
– Read 5 CHANNEL_STATUS_1 and verify the following bits:
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
77
TLK6002
SLLSE34 – MAY 2010
•
•
•
•
•
•
•
•
•
78
www.ti.com
5.14 AZ_DONE (1’b1) (per channel)
5.13 AGC_LOCKED (1’b1) (per channel)
5.7 TX_FIFO_UNDERFLOW (1’b0) (per channel)
5.6 TX_FIFO_OVERFLOW (1’b0) (per channel)
5.5 RX_FIFO_UNDERFLOW (1’b0) (per channel)
5.4 RX_FIFO_OVERFLOW (1’b0) (per channel)
5.2 LOS (1’b0) (per channel). This read value is only useful if the serial input is guaranteed to
be above 150mVdfpp.
5.1 CHANNEL_SYNC (1’b1) (per channel)
5.0 PLL_LOCK (1’b1) (per channel)
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
4.18.2 16-Bit Interface Mode (8b/10b Encoder/Decoder Enabled) (All CPRI/OBSAI Rates)
Note: Assume both channel A and channel B have the same setup.
REFCLK frequency = 122.88 MHz, Mode = Transceiver, Parallel Interface = 16-Bit DDR Source Aligned
Mode, RXCLK_A/B out = RXBCLK_A/B, Serial Data Rate is CPRI/OBSAI standard rate as shown below.
• Device Pin Setting(s) – Pin settings allow for maximum software configurability.
– Ensure CODEA_EN input pin is Low.
– Ensure CODEB_EN input pin is Low.
– Ensure RATE_A[2:0] input pins are 3’b100 (High, Low, Low) to enable software control.
– Ensure RATE_B[2:0] input pins are 3’b100 (High, Low, Low) to enable software control.
– Ensure PD_TRXA_N input pin is High.
– Ensure PD_TRXB_N input pin is High.
– Ensure PRBS_EN input pin is Low.
– Ensure REFCLK_A_SEL input pin is Low to enable software control.
– Ensure REFCLK_B_SEL input pin is Low to enable software control.
• Reset Device
– Issue a hard or soft reset (RESET_N asserted for at least 10 us -or- Write 1’b1 to 0.15
GLOBAL_RESET) after power supply stabilization.
• Enable MDIO global write so that each MDIO write affects both channels to shorten provisioning time
– Write 1’b1 to 0.11 GLOBAL_WRITE
• Clock Configuration
– Select Channel A SERDES REFCLK input (Default = REFCLK_0_P/N)
• If REFCLK_0_P/N used – Write 1’b0 to 0.1 REFCLK_A_SEL
• If REFCLK_1_P/N used – Write 1’b1 to 0.1 REFCLK_A_SEL
– Select Channel B SERDES REFCLK input (Default = REFCLK_0_P/N)
• If REFCLK_0_P/N used – Write 1’b0 to 0.0 REFCLK_B_SEL
• If REFCLK_1_P/N used – Write 1’b1 to 0.0 REFCLK_B_SEL
• Data Rate Setting (select one of the following 8 cases)
– If serial data rate is 6144.00Mbps: Write 2’b00 to 1.7:6 RATE_TX[1:0], write 2’b00 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select FULL rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x010E).
– If serial data rate is 4915.20Mbps: Write 2’b00 to 1.7:6 RATE_TX[1:0], write 2’b00 to 1.5:4
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select FULL rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x010D).
– If serial data rate is 3072.00Mbps: Write 2’b01 to 1.7:6 RATE_TX[1:0], write 2’b01 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select HALF rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x015E).
– If serial data rate is 2457.60Mbps: Write 2’b01 to 1.7:6 RATE_TX[1:0], write 2’b01 to 1.5:4
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select HALF rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x015D).
– If serial data rate is 1536.00Mbps: Write 2’b10 to 1.7:6 RATE_TX[1:0], write 2’b10 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select QUARTER rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x01AE).
– If serial data rate is 1228.80Mbps: Write 2’b10 to 1.7:6 RATE_TX[1:0], write 2’b10 to 1.5:4
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select QUARTER rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x01AD).
– If serial data rate is 768.00Mbps: Write 2’b11 to 1.7:6 RATE_TX[1:0], write 2’b11 to 1.5:4
RATE_RX[1:0], write 4’b1110 to 1.3:0 PLL_MULT[3:0] to select EIGHTH rate and 25x MPY
(CHANNEL_CONTROL_1 = 0x01FE).
– If serial data rate is 614.40Mbps: Write 2’b11 to 1.7:6 RATE_TX[1:0], write 2’b11 to 1.5:4
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
79
TLK6002
SLLSE34 – MAY 2010
•
•
•
•
•
•
•
•
•
•
80
www.ti.com
RATE_RX[1:0], write 4’b1101 to 1.3:0 PLL_MULT[3:0] to select EIGHTH rate and 20x MPY
(CHANNEL_CONTROL_1 = 0x01FD).
Serial Configuration
– Configure the following bits per the desired application
• 1.9:8 (LOOP_BANDWIDTH[1:0])
• 2.12:8 (TWPOST1[4:0])
• 2.7:4 (TWPRE[3:0])
• 2.3:0 (SWING[3:0])
• 8.14:12 (EQPRE[2:0])
• 8.11:10 (CDRTHR[1:0]) = 2’b01
• 8.9:8 (CDRFMULT[1:0]) = 2’b00
Mode Control
– Channel synchronization (comma enable) is on by default and the parallel output 9 bit codes are
byte aligned.
– Write 1’b1 to 3.3 ENCODE_ENABLE
– Write 1’b1 to 3.2 DECODE _ENABLE
– Set the DDR Source Aligned Mode
• Write 1’b1 to 3.6 DDR_ENABLE
• Write 1’b1 to 3.1 TX_EDGE_MODE to select Transmit DDR Source Aligned Mode
• Write 1’b1 to 3.0 RX_EDGE_MODE to select Receive DDR Source Aligned Mode
If a different parallel IO align mode used:
– If SDR Falling Edge Aligned: write 0x018C to CHANNEL_CONTROL_3 register.
– If SDR Rising Edge Aligned: write 0x018F to CHANNEL_CONTROL_3 register.
– If DDR Source Centered: write 0x01CC to CHANNEL_CONTROL_3 register.
Enable desired status signals to LOSA and LOSB for real time monitoring per channel. Any number of
signals can be enabled at once.
– If SERDES Rx Loss of Signal condition monitored: write 1’b1 to 6.10 LOS_OVERLAY.
– If channel synchronization status monitored: write 1’b1 to 6.9 CH_SYNC_OVERLAY.
– If PLL lock status monitored: write 1’b1 to 6.8 PLL_LOCK_OVERLAY.
– If invalid code word status monitored: write 1’b1 to 6.3 INVALID_CODE_OVERLAY
– If SERDES AGC unlock status monitored: write 1’b1 to 7.7 AGCLOCK_OVERLAY.
– If SERDES AZDONE status monitored: write 1’b1 to 7.6 AZDONE_OVERLAY.
Check SERDES PLL Status for Locked State
– Poll 5.0 PLL_LOCK (per channel) until it is asserted (high).
Toggle ENRX
– Write 1’b0 to 20.2 (ENRX)
– Write 1’b1 to 20.2 (ENRX)
Final CDR Configuration
– Wait until either AGC_LOCKED asserted or 250M UI
– Write 8.9:8 (CDRFMULT[1:0]) = 2’b01
– Poll 5.13 AGC_LOCKED (1’b1) (per channel)
Issue Data path Reset
– Write 1’b1 to 4.3 DATAPATH_RESET
Clear Latched Registers
– Read 5 CHANNEL_STATUS_1 to clear all (per channel)
– Read 14.15:0 ERROR_COUNTER to clear (per channel)
Device provisioning has completed at this point
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
•
SLLSE34 – MAY 2010
Periodically Check Device Operational Mode Status (Non-Errored Read Values Shown Below):
– Read 5 CHANNEL_STATUS_1 and verify the following bits:
• 5.14 AZ_DONE (1’b1) (per channel)
• 5.13 AGC_LOCKED (1’b1) (per channel)
• 5.9 ENCODE_INVALID (1’b0) (per channel)
• 5.8 DECODE_INVALID (1’b0) (per channel)
• 5.7 TX_FIFO_UNDERFLOW (1’b0) (per channel)
• 5.6 TX_FIFO_OVERFLOW (1’b0) (per channel)
• 5.5 RX_FIFO_UNDERFLOW (1’b0) (per channel)
• 5.4 RX_FIFO_OVERFLOW (1’b0) (per channel)
• 5.2 LOS (1’b0) (per channel). This read value is only useful if the serial input is guaranteed to
be above 150mVdfpp.
• 5.1 CHANNEL_SYNC (1’b1) (per channel)
• 5.0 PLL_LOCK (1’b1) (per channel)
– Read 14.15:0 ERROR_COUNTER[15:0] and verify it is 0 (per channel)
4.19 JITTER TEST PATTERN GENERATION AND VERIFICATION PROCEDURES
Use one of the following procedures to generate and verify the respective test patterns. It is assumed that
an appropriate external cable has been connected between serial outputs and serial inputs. No external
parallel side connections are necessary.
4.19.1 IEEE802.3 Clause 36A Based Continuous Random Pattern (CRPAT) Long/Short
Test Pattern:
•
•
•
•
•
•
Reset Device:
– Issue a hard or soft reset (RESET_N asserted –or- Write 1 to 0.15 GLOBAL_RESET)
Select SERDES Reference Clock Input:
– If REFCLK_0_P/N used - Ensure REFCLK_A_SEL (or REFCLK_B_SEL if channel B is used)
primary input pin is low
– If REFCLK_1_P/N used - Ensure REFCLK_A_SEL (or REFCLK_B_SEL if channel B is used)
primary input pin is high
Ensure a legal reference clock operation frequency is selected based on Appendix B (Continuous Rate
Device Configuration), and provision CHANNEL_CONTROL_1 register accordingly.
Serial Configuration
– Configure the following bits per the desired application:
• 1.9:8 (LOOP_BANDWIDTH[1:0])
• 2.12:8 (TWPOST1[4:0])
• 2.7:4 (TWPRE[3:0])
• 2.3:0 (SWING[3:0])
• 8.14:12 (EQPRE[2:0])
• 8.11:10 (CDRTHR[1:0]) = 2’b01
• 8.9:8 (CDRFMULT[1:0]) = 2’b00
• 1.3:0 PLL_MULT[3:0]
• 1.7:6 RATE_TX[1:0] (CRPAT supported only in half/quarter/eighth rate modes
• 1.5:4 RATE_RX[1:0] (CRPAT supported only in half/quarter/eighth rate modes).
Check SERDES PLL Status for Locked State
– Poll 5.0 PLL_LOCK (per channel) until it is asserted (high).
Toggle ENRX
– Write 1’b0 to 20.2 (ENRX)
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
81
TLK6002
SLLSE34 – MAY 2010
•
•
•
•
•
•
•
•
•
www.ti.com
– Write 1’b1 to 20.2 (ENRX)
Select Test Pattern:
– If CRPAT Long Pattern is desired:
• Write 3’b011 to 7.10:8 TEST_PATTERN_SEL[2:0]
– If CRPAT Short Pattern is desired:
• Write 3’b100 to 7.10:8 TEST_PATTERN_SEL[2:0]
Enable Test Pattern Generation:
– Write 1’b1 to 7.13 TP_GEN_EN
– Poll 5.13 AGC_LOCKED (1’b1) (per channel)
Final CDR Configuration
– Wait until either AGC_LOCKED asserted or 250M UI
– Write 8.9:8 (CDRFMULT[1:0]) = 2’b01
– Poll 5.13 AGC_LOCKED (1’b1) (per channel)
Issue Data path Reset
– Write 1’b1 to 4.3 DATAPATH_RESET
Enable Test Pattern Verification:
– Write 1’b1 to 7.12 TP_VERIFY_EN
Clear Counters:
– Read 14.15:0 ERROR_COUNTER[15:0] and discard the value.
Verify Test In Progress:
– Poll 5.12 TP_STATUS until asserted.
The pattern verification is now in progress.
Verify Error Free Operation (as many times as desired during the duration of the test period):
– Read 14.15:0 ERROR_COUNTER[15:0], and verify 16’h0000 is read to confirm error free
operation.
If more than one test is specified results are unpredictable.
If another test type is desired, begin at the first step of that procedure.
82
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
4.19.2 PRBS TEST GENERATION AND VERIFICATION PROCEDURES
Use one of the following procedures to generate and verify the respective PRBS test patterns. It is
assumed that an appropriate external cable has been connected between serial outputs and serial inputs.
No external parallel side connections are necessary.
4.19.2.1 27-1 / 223-1 / 231-1 PRBS Register Based Testing
•
•
•
•
•
•
•
•
•
•
•
Note: PRBS TX does not support eighth rate mode.
Reset Device:
– Issue a hard or soft reset (RESET_N asserted -or- Write 1 to 0.15 GLOBAL_RESET)
Select SERDES Reference Clock Input:
– If REFCLK_0_P/N used – Ensure REFCLK_A_SEL (or REFCLK_B_SEL if channel B is used)
primary input pin is low
– If REFCLK_1_P/N used – Ensure REFCLK_A_SEL (or REFCLK_B_SEL if channel B is used)
primary input pin is high
Ensure a legal reference clock operation frequency is selected based on Appendix B (Continuous Rate
Device Configuration), and provision CHANNEL_CONTROL_1 register accordingly. (Note: Eighth
Rate TX Is Not Supported).
Serial Configuration
– Configure the following bits per the desired application:
• 1.9:8 (LOOP_BANDWIDTH[1:0])
• 2.12:8 (TWPOST1[4:0])
• 2.7:4 (TWPRE[3:0])
• 2.3:0 (SWING[3:0])
• 8.14:12 (EQPRE[2:0])
• 8.11:10 (CDRTHR[1:0]) = 2’b01
• 8.9:8 (CDRFMULT[1:0]) = 2’b00
• 1.3:0 PLL_MULT[3:0]
• 1.7:6 RATE_TX[1:0]
• 1.5:4 RATE_RX[1:0]
Check SERDES PLL Status for Locked State
– Poll 5.0 PLL_LOCK (per channel) until it is asserted (high).
Toggle ENRX
– Write 1’b0 to 20.2 (ENRX)
– Write 1’b1 to 20.2 (ENRX)
Select Test Pattern:
– If 27-1 PRBS Pattern is desired:
• Write 3’b101 to 7.10:8 TEST_PATTERN_SEL[2:0]
– If 223-1 PRBS Pattern is desired:
• Write 3’b110 to 7.10:8 TEST_PATTERN_SEL[2:0]
– If 231-1 PRBS Pattern is desired:
• Write 3’b111 to 7.10:8 TEST_PATTERN_SEL[2:0]
Enable Test Pattern Generation:
– Write 1’b1 to 7.13 TP_GEN_EN
Final CDR Configuration
– Wait until either AGC_LOCKED asserted or 250M UI
– Write 8.9:8 (CDRFMULT[1:0]) = 2’b01
– Poll 5.13 AGC_LOCKED (1’b1) (per channel)
Issue Data path Reset
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
83
TLK6002
SLLSE34 – MAY 2010
•
•
•
•
•
www.ti.com
– Write 1’b1 to 4.3 DATAPATH_RESET
Enable Test Pattern Verification:
– Write 1’b1 to 7.12 TP_VERIFY_EN
The pattern verification is now in progress.
PRBSA_PASS contains a real time output that when low indicates the input PRBS pattern on RXAP/N
contains error.
PRBSB_PASS contains a real time output that when low indicates the input PRBS pattern on RXBP/N
contains error.
The following steps can be performed if the number of errors needs to be monitored:
– Read 14.15:0 ERROR_COUNTER[15:0] and discard the value.
– Read 14.15:0 ERROR_COUNTER[15:0], and verify 16’h0000 is read to confirm error free
operation. (as many times as desired during the duration of the test period)
4.19.2.2 27-1 / 223-1 / 231-1 PRBS Pin Based Testing
•
•
•
•
•
•
•
•
•
84
Note: PRBS TX does not support eighth rate mode.
Device Pin Setting(s):
– Ensure PRBS_EN primary input pin is high.
– PRBS Selection
• PRBS 231-1 will be selected by default.
Reset Device:
– Issue a hard or soft reset (RESET_N asserted –or- Write 1 to 0.15 GLOBAL_RESET)
Select SERDES Reference Clock Input:
– If REFCLK_0_P/N used - Ensure REFCLK_A_SEL (or REFCLK_B_SEL if channel B is used)
primary input pin is low
– If REFCLK_1_P/N used - Ensure REFCLK_A_SEL (or REFCLK_B_SEL if channel B is used)
primary input pin is high
Ensure a legal reference clock operation frequency is selected based on Appendix B (Continuous Rate
Device Configuration), and provision CHANNEL_CONTROL_1 register accordingly. (Note: Eighth
Rate TX Is Not Supported).
Serial Configuration
– Configure the following bits per the desired application:
• 1.9:8 (LOOP_BANDWIDTH[1:0])
• 2.12:8 (TWPOST1[4:0])
• 2.7:4 (TWPRE[3:0])
• 2.3:0 (SWING[3:0])
• 8.14:12 (EQPRE[2:0])
• 8.11:10 (CDRTHR[1:0]) = 2’b01
• 8.9:8 (CDRFMULT[1:0]) = 2’b00
• 1.3:0 PLL_MULT[3:0]
• 1.7:6 RATE_TX[1:0]
• 1.5:4 RATE_RX[1:0]
Check SERDES PLL Status for Locked State
– Poll 5.0 PLL_LOCK (per channel) until it is asserted (high).
Toggle ENRX
– Write 1’b0 to 20.2 (ENRX)
– Write 1’b1 to 20.2 (ENRX)
Select Test Pattern if either 27-1 or 223-1 PRBS Pattern is desired. Skip this step if 231-1 PRBS
Pattern is desired.
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
•
•
•
•
•
•
SLLSE34 – MAY 2010
– If 27-1 PRBS Pattern is desired:
• Write 3’b101 to 7.10:8 TEST_PATTERN_SEL[2:0]
– If 223-1 PRBS Pattern is desired:
• Write 3’b110 to 7.10:8 TEST_PATTERN_SEL[2:0]
Final CDR Configuration
– Wait until either AGC_LOCKED asserted or 250M UI
– Write 8.9:8 (CDRFMULT[1:0]) = 2’b01
– Poll 5.13 AGC_LOCKED (1’b1) (per channel)
Issue Data path Reset
– Write 1’b1 to 4.3 DATAPATH_RESET
The pattern verification is now in progress.
PRBSA_PASS contains a real time output that when low indicates the input PRBS pattern on RXAP/N
contains error.
PRBSB_PASS contains a real time output that when low indicates the input PRBS pattern on RXBP/N
contains error.
The following steps can be performed if the number of errors needs to be monitored:
– Read 14.15:0 ERROR_COUNTER[15:0] and discard the value.
– Read 14.15:0 ERROR_COUNTER[15:0], and verify 16’h0000 is read to confirm error free
operation. (as many times as desired during the duration of the test period)
4.20 PACKAGE DISSIPATION RATINGS
The following tables detail the thermal characteristics of the TLK6002 package.
Table 4-2. Package Thermal Characteristics
PARAMETER
JEDEC
STANDARD BOARD
CUSTOM TYPICAL
APPLICATION BOARD
VALUE
VALUE
qJA
Theta-JA
24.3
19.2
qJB
Theta-JB
11.6
11.6
qJC
Theta-JC
4.5
4.5
ΨJT
Psi-JT
4.5
4.5
ΨJB
Psi-JB
11.5
11.3
Note: Custom Typical Application Board Characteristics:
• 10 x 15 inches
• 12 layer
– 8 power/ground layers – 95% copper (1oz)
– 4 signal layers – 20% copper (1oz)
ΨJB = (TJ – TB)/(Total Device Power Dissipation)
TJ = Device Junction Temperature
TB = Temperature of PCB 1mm from device edge.
ΨJT = (TJ – TC)/(Total Device Power Dissipation)
TJ = Device Junction Temperature
TC = Hottest temperature on the case of the package
4.21 Device Total Worst Case Power Dissipation
The following table details the worst case total device power dissipation:
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
85
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 4-3. Device Worst Case Total Power Dissipation
VDDQA/B, VDDRA/B, VDDO1/2/3 = 1.6V
VDDQA/B, VDDRA/B, VDDO1/2/3 = 1.9V
HSTL Input Termination
Serial Rate
(Gbps)
Disabled
(6.1:0 = 2'b00)
HSTL Input Termination
Half Strength
(6.1:0 = 2'b01)
Full Strength
(6.1:0 = 2'b1x)
Disabled
(6.1:0 = 2'b00)
Half Strength
(6.1:0 = 2'b01)
Full Strength
(6.1:0 = 2'b1x)
0.47
552
994
1322
674
1301
1775
0.6144
596
1024
1351
727
1348
1813
0.768
653
1067
1395
781
1395
1864
1.2288
677
1065
1371
834
1389
1842
1.536
750
1129
1443
906
1462
1889
2.4576
829
1162
1440
1032
1498
1898
3.072
917
1225
1499
1131
1574
1954
4.9512
975
1245
1507
1210
1612
1952
6.144
1046
1315
1562
1305
1690
2038
6.25
1057
1320
1580
1286
1696
2064
The test conditions for above data are: two channels active in transceiver mode, parallel/serial loopback
connections installed external to the device, CRPAT Long data pattern, all unlisted power supplies at
1.05V, and T=85C.
Device Errata:
If VDDQ power supply ramp leads DVDD, the following items must be noted:
1. Register 0x0F usage is not possible
2. CS_N/GPI1/SDI/SDO/DCI signals are not usable
3. Register 28.14 must be asserted manually for CLK_OUT_P/N normal operation
86
ELECTRICAL SPECIFICATIONS
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
5 Appendix A – Application Board Supply Recommendations
Nominal Board Voltages Required: 1.0V, and either 1.5 or 1.8V
Ground - Connect all device grounds together on the application board (DGND, AGND)
1.0V – Bulk Decoupled
1. Ferrite Bead → DVDD (Locally Decoupled)
2. Ferrite Bead → AVDD / VDDT / VDDD (Locally Decoupled)
1.5V or 1.8V – Bulk Decoupled
1. Ferrite Bead → VDDQA/B (Locally Decoupled)
2. Ferrite Bead → VDDRA,VDDRB (Separate, Locally Decoupled)
3. Ferrite Bead → VDDO1,VDDO2,VDDO3 (Locally Decoupled)
VREF* can be generated using 1k resistors between VDDQA/B and DGND, and should be locally
decoupled.
Note: Ferrite beads should have high impedance near the fundamental frequency that the parallel data is
running at.
Note: Bulk decoupling prior to the ferrite beads required in order to decouple noise from power source.
This decoupling will be determined by your power source and what frequencies of noise it creates.
Note: For close to device localized decoupling (after ferrite beads), 1.0µF and 0.1µF caps should be
placed on the pins of the device on the back side of the board.
It is strongly recommended that TI review relevant pages of your application board schematics and layout
before fabrication to ensure first pass success.
Copyright © 2010, Texas Instruments Incorporated
Appendix A – Application Board Supply Recommendations
Submit Documentation Feedback
Product Folder Link(s): TLK6002
87
TLK6002
SLLSE34 – MAY 2010
www.ti.com
6 Appendix B – Continuous Rate Device Configuration
The REFCLK needed based on a particular SERDES configuration is as follows:
SERDES Reference Clock = (Rate Scale) × (Serial Bit Rate) / (Serdes Multiplier)
Note that Table 6-1 indicates legal ranges for each setting.
Table 6-1. Continuous Rate SERDES Configuration Settings
SERDES Serial Rate Configuration
SERDES Reference Clock
SERDES
Multiplier
Period (ns)
Serial Rate (Gbps)
Frequency
(MHz)
Full
Half
Quarter
Eighth
Min
Max
Max
Min
Max
Max
Min
Max
Max
Min
Max
Max
4
1.25
2.14
800
467
3.74
6.25
1.87
3.20
0.93
1.60
0.47
0.80
5
1.33
2.67
752
375
3.75
6.25
1.87
3.76
0.94
1.88
0.47
0.94
6
1.6
3.2
625
313
3.75
6.25
1.88
3.75
0.94
1.88
0.47
0.94
8
2.13
4.26
469
235
3.76
6.25
1.88
3.76
0.94
1.88
0.47
0.94
8.25
2.19
4.4
457
227
3.75
6.25
1.88
3.77
0.94
1.88
0.47
0.94
10
2.67
5.33
375
188
3.75
6.25
1.88
3.75
0.94
1.87
0.47
0.94
12
3.2
6.4
313
156
3.75
6.25
1.88
3.75
0.94
1.88
0.47
0.94
15
4
8.14
250
122.88
3.69
6.25
1.84
3.75
0.92
1.88
0.46
0.94
16
4.25
8.14
235
122.88
3.93
6.25
1.97
3.76
0.98
1.88
0.49
0.94
16.5
4.39
8.14
228
122.88
4.06
6.25
2.03
3.76
1.01
1.88
0.51
0.94
20
5.33
8.14
188
122.88
4.92
6.25
2.46
3.75
1.23
1.88
0.61
0.94
25
6.67
8.14
150
122.88
6.14
6.25
3.07
3.75
1.54
1.87
0.77
0.94
Note: In Full Rate Mode, the SERDES Reference Clock range shown above must be limited such that serial side operation does not exceed
6.25 Gbps.
Rate Scale
Full
Half
Quarter
Eighth
0.5
1
2
4
Serial performance is optimal when the highest reference clock and lowest SERDES multiplier is chosen
for a given application rate.
88
Appendix B – Continuous Rate Device Configuration
Submit Documentation Feedback
Product Folder Link(s): TLK6002
Copyright © 2010, Texas Instruments Incorporated
TLK6002
www.ti.com
SLLSE34 – MAY 2010
7 Appendix C – 8b/10B Control Characters Supported
Table 7-1 contains 8b/10b control characters are valid and are supported in TLK6002.
The K28.7 usage is not advised as comma characters can be created on non-symbol boundaries without
depending upon adjacent characters.
Further detail on 8b/10b encoding/decoding of data characters can be found in IEEE802.3-2002 Clause
36.
Table 7-1. 8b/10b Control Characters Supported
CONTROL
CHARACTER
DATA BYTE
K28.0
0x1C
K28.1
0x3C
K28.2
0x5C
K28.3
0x7C
K28.4
0x9C
K28.5
0xBC
K28.6
0xDC
K28.7
0xFC
K23.7
0xF7
K27.7
0xFB
K29.7
0xFD
K30.7
0xFE
Copyright © 2010, Texas Instruments Incorporated
CONTAINS COMMA
CHARACTER
Y
Y
Y
Appendix C – 8b/10B Control Characters Supported
Submit Documentation Feedback
Product Folder Link(s): TLK6002
89
TLK6002
SLLSE34 – MAY 2010
www.ti.com
8 Appendix D – Device Latency Specification
The following tables show the absolute device latency in each operation mode.
Table 8-1. Device Absolute Transmit Latency – SDR and DDR Modes
SDR TRANSMIT ABSOLUTE LATENCY (Serial Bit Times)
Note: TX FIFO, in register 6.14:12, defaults to Auto Selection Mode. This table contains latencies when the TX FIFO is set in the minimum
latency mode (6.14:12 = 3'b000).
LATENCY COMPONENTS
Item
Minimum
Maximum
A
8
12
Sampling edge of TXCLK_* to TX FIFO Input Register
B
10
40
TX FIFO Input Register to TX FIFO Output Register
C
20
20
TX FIFO Output Register to 8b/10b Encoder Register
D
10
10
8b/10b Encoder Register to PMA Retime Register
E
20
20
PMA Retime Register to TX SERDES Input Register
F
16
22
TX SERDES Input Register to Serialized Output Bit
Bit Detail
Latency
Minimum
Maximum
TX*_[19]
0
84
124
TX*_[10]
9
93
133
TX*_[9]
10
94
134
TX*_[0]
19
103
143
Latency
Summary
Encoder Enabled
Description
Assumption: TX*_[19] is first transmitted bit. TX*_[19:10] is first transmitted
symbol. 8b/10b encoder is enabled. Disabling 8b/10b encoder reduces latency
by 10 bit times.
Encoder Disabled
Minimum
Maximum
Minimum
Maximum
TX*_[19]
84
124
74
114
TX*_[10]
93
133
83
123
TX*_[9]
94
134
84
124
TX*_[0]
103
143
93
133
DDR TRANSMIT ABSOLUTE LATENCY (Serial Bit Times)
DDR Transmit Latency is the same as transmit SDR mode, except that all numbers are 8 to12 bit times larger.
Latency
Summary
90
Encoder Enabled
Encoder Disabled
Minimum
Maximum
Minimum
Maximum
TX*_[19]
92
136
82
126
TX*_[10]
101
145
91
135
TX*_[9]
102
146
92
136
TX*_[0]
111
155
101
145
Appendix D – Device Latency Specification
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
Table 8-2. Device Absolute Receive Latency – SDR and DDR Modes
RX SDR/DDR RECEIVE ABSOLUTE LATENCY
Note: RX FIFO Depth (When FIFO Enabled) is 4 words.
Latency Components
Item
Minimum
Maximum
A
47
57
Serial Input Bit to RX SERDES Output Register
Description
B
10
10
RX SERDES Output Register to RX PMA Register
C
10
10
RX PMA Register to Channel Sync Input Register
D
10
20
Channel Sync Input Register to Channel Sync Output Register
E
10
10
Channel Sync Output Register to 8b/10b Decoder Input Register
F
10
10
8b/10b Decoder Input Register to 8b/10b Output Register
G
10
10
8b/10b Output Register to RX FIFO Input Register
H
10
40
RX FIFO Input Register to RX FIFO Output Register
I
10
20
RXFIFO Output Register to RX Parallel Output Register Driven
Bit Detail
Latency
Minimum
Maximum
RX*_[19]
0
127
187
RX*_[10]
9
136
196
RX*_[9]
0
127
187
RX*_[0]
9
136
196
Assumption: RX*_[10 and 0] are the first received bits.
Disabling Channel Synchronization Reduces Above Numbers by 20 Minimum and 30 Maximum bit times.
Disabling Decoder Reduces Above Numbers by 20 bit times.
Bypassing RX FIFO Reduces Above Numbers by 20 Minimum and 50 Maximum bit times.
Latency
Summary
RX FIFO
Bypassed
RX FIFO
Bypassed
RX FIFO
Bypassed
RX FIFO
Enabled
Channel
Sync On
Channel Sync
On
Channel
Sync Off
Channel
Sync On
8b/10b
Decoder On
8b/10b Decoder
Off
8b/10b
Decoder Off
8b/10b
Decoder On
Minimum
Maximum
Minimum
Maximum
Minimum
Maximum
Minimum
Maximum
RX*_[19]
107
137
87
117
67
87
127
187
RX*_[10]
116
146
96
126
76
96
136
196
RX*_[9]
107
137
87
117
67
87
127
187
RX*_[0]
116
146
96
126
76
96
136
196
Table 8-3 show the static latency variance. Note that static latency variation is the difference in absolute
latency values possible across device reset/power ups.
Appendix D – Device Latency Specification
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
91
TLK6002
SLLSE34 – MAY 2010
www.ti.com
Table 8-3. Device Static Latency Variance
TRANSMIT STATIC LATENCY VARIANCE – Maximum
8b/10b Encoder
Interface Mode
Serial Bit Times
Enabled
SDR
33
Enabled
DDR
33
Disabled
SDR
33
Disabled
DDR
33
Note: TX_FIFO_DEPTH[2:0] = 3'b000
RECEIVE STATIC LATENCY VARIANCE – Maximum
Receive FIFO
Channel
Synchronization
8b/10b Decoder
Serial Bit Times
Bypassed
Bypassed
Enabled
Enabled
30
Enabled
Disabled
30
Bypassed
Disabled
Disabled
20
Enabled
Enabled
Enabled
60
Table 8-4 shows the device dynamic latency variance. Note that the dynamic latency variation is the
difference in latency when voltage and temperature are varied for a particular absolute static latency, after
traffic (including channel synchronization) has been established. The dynamic latency variance numbers
do not include phase movement between the parallel input clocks and input reference clocks.
Table 8-4. Device Dynamic Latency Variance
Transmit Dynamic Latency Variance
92
Receive Dynamic Latency Variance
Minimum
Maximum
Minimum
Maximum
0 ns
2 ns
0 ns
2 ns
Appendix D – Device Latency Specification
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
TLK6002
www.ti.com
SLLSE34 – MAY 2010
9 Appendix E – Device Test Modes
This device can be placed into one of the four modes: functional mode including JTAG testing mode, scan
1 testing mode, scan 2 testing mode, and Cooper/eFuse testing mode. The scan 1, scan 2 and
Cooper/eFuse testing modes are for TI use only, and may be ignored by external users of this device.
Table 9-1. Device Mode Configuration
FUNCTIONAL DEVICE
PIN NAME
FUNCTIONAL MODE/JTAG
TESTING
SCAN 1 MODE
SCAN 2 MODE
COOPER/EFUSE MODE
TESTEN
0
0
1
1
GPI0
0
1
0
1
Table 9-2. Device Test Mode Pin Configuration
FUNCTIONAL
DEVICE PIN
NAME
FUNCTIONAL
MODE SIGNAL
DIRECTION
TEST MODE
SIGNAL
DIRECTION
FUNCTIONAL
MODE/JTAG
TESTING
CODEA_EN
I
I
CODEA_EN
Scan In 8
Scan In 8
STCICFG0
RATE_A2
I
I
RATE_A2
Scan In 7
Scan In 7
STCI_D
RATE_A1
I
I
RATE_A1
Scan In 6
Scan In 6
EFUSE_TMS
RATE_A0
I
I
RATE_A0
Scan In 5
Scan In 5
EFUSE_TDI
RATE_B2
I
I
RATE_B2
Scan In 4
Scan In 4
STCICFG1
RATE_B1
I
I
RATE_B1
Scan In 3
Scan In 3
EFUSE_INITZ
RATE_B0
I
I
RATE_B0
Scan In 2
Scan In 2
EFUSE_TRST
REFCLK_A_SEL
I
I
REFCLK_A_SEL
Scan In 1
Scan In 1
REFCLK_A_SEL
REFCLK_B_SEL
I
I
REFCLK_B_SEL
Scan Enable
Scan Enable
REFCLK_B_SEL
CLK_OUT_SEL
I
I
CLK_OUT_SEL
At Speed Scan Enable
0: stuck-at fault
1: transition fault
HSTL Force Up
EFUSE_TCK
CODEB_EN
I
I
CODEB_EN
Scan Clock Select (0:
from device pin, 1: from
Cooper), also
EFUSE_SYS_CLK
HSTL Force Down
EFUSE_SYS_CLK
TDI
I
I
TDI
Adaptive Scan Enable
(Test Mode)
Adaptive Scan
Enable (Test Mode)
TDI
PRBS_EN
I
I
PRBS_EN
Scan Clock
Scan Clock
STCICLK
PRTAD4
I
O or I
PRTAD4
Scan Out 8 (O)
Scan Out 8 (O)
TESTCLK_R (I)
PRTAD3
I
O or I
PRTAD3
Scan Out 7 (O)
Scan Out 7 (O)
efuse_cooper_sel (I)
(0: Cooper mode, 1:
eFuse mode)
PRTAD2
I
O or I
PRTAD2
Scan Out 6 (O)
Scan Out 6 (O)
TESTCLK_T (I)
PRTAD1
I
O or I
PRTAD1
Scan Out 5 (O)
Scan Out 5 (O)
Not Used (I)
PRTAD0
I
O or I
PRTAD0
Scan Out 4 (O)
Scan Out 4 (O)
Not Used (I)
PRBSA_PASS
O
O
PRBSA_PASS
Scan Out 3 (O)
Scan Out 3 (O)
Tied Low
LOSA
O
O
LOSA
Scan Out 2 (O)
Scan Out 2 (O)
STCI_Q
PRBSB_PASS
O
O
PRBSB_PASS
Scan Out 1 (O)
Scan Out 1 (O)
EFUSE_TDO
LOSB
O
O
LOSB
Burnin_Output
Burnin_Output
Burnin_Output
SCAN 1 MODE
SCAN 2 MODE
COOPER/EFUSE
MODE
Appendix E – Device Test Modes
Copyright © 2010, Texas Instruments Incorporated
Submit Documentation Feedback
Product Folder Link(s): TLK6002
93
PACKAGE OPTION ADDENDUM
www.ti.com
14-May-2010
PACKAGING INFORMATION
Orderable Device
Status (1)
Package
Type
Package
Drawing
TLK6002ZEU
ACTIVE
BGA
ZEU
Pins Package Eco Plan (2)
Qty
324
84
TBD
Lead/Ball Finish
Call TI
MSL Peak Temp (3)
Call TI
(1)
The marketing status values are defined as follows:
ACTIVE: Product device recommended for new designs.
LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect.
NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in
a new design.
PREVIEW: Device has been announced but is not in production. Samples may or may not be available.
OBSOLETE: TI has discontinued the production of the device.
(2)
Eco Plan - The planned eco-friendly classification: Pb-Free (RoHS), Pb-Free (RoHS Exempt), or Green (RoHS & no Sb/Br) - please check
http://www.ti.com/productcontent for the latest availability information and additional product content details.
TBD: The Pb-Free/Green conversion plan has not been defined.
Pb-Free (RoHS): TI's terms "Lead-Free" or "Pb-Free" mean semiconductor products that are compatible with the current RoHS requirements
for all 6 substances, including the requirement that lead not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered
at high temperatures, TI Pb-Free products are suitable for use in specified lead-free processes.
Pb-Free (RoHS Exempt): This component has a RoHS exemption for either 1) lead-based flip-chip solder bumps used between the die and
package, or 2) lead-based die adhesive used between the die and leadframe. The component is otherwise considered Pb-Free (RoHS
compatible) as defined above.
Green (RoHS & no Sb/Br): TI defines "Green" to mean Pb-Free (RoHS compatible), and free of Bromine (Br) and Antimony (Sb) based flame
retardants (Br or Sb do not exceed 0.1% by weight in homogeneous material)
(3)
MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder
temperature.
Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is
provided. TI bases its knowledge and belief on information provided by third parties, and makes no representation or warranty as to the
accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken and continues to take
reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on
incoming materials and chemicals. TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited
information may not be available for release.
In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI
to Customer on an annual basis.
Addendum-Page 1
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications, enhancements, improvements,
and other changes to its products and services at any time and to discontinue any product or service without notice. Customers should
obtain the latest relevant information before placing orders and should verify that such information is current and complete. All products are
sold subject to TI’s terms and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in accordance with TI’s standard
warranty. Testing and other quality control techniques are used to the extent TI deems necessary to support this warranty. Except where
mandated by government requirements, testing of all parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for their products and
applications using TI components. To minimize the risks associated with customer products and applications, customers should provide
adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right, copyright, mask work right,
or other TI intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information
published by TI regarding third-party products or services does not constitute a license from TI to use such products or services or a
warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual
property of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of TI information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied
by all associated warranties, conditions, limitations, and notices. Reproduction of this information with alteration is an unfair and deceptive
business practice. TI is not responsible or liable for such altered documentation. Information of third parties may be subject to additional
restrictions.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that product or service voids all
express and any implied warranties for the associated TI product or service and is an unfair and deceptive business practice. TI is not
responsible or liable for any such statements.
TI products are not authorized for use in safety-critical applications (such as life support) where a failure of the TI product would reasonably
be expected to cause severe personal injury or death, unless officers of the parties have executed an agreement specifically governing
such use. Buyers represent that they have all necessary expertise in the safety and regulatory ramifications of their applications, and
acknowledge and agree that they are solely responsible for all legal, regulatory and safety-related requirements concerning their products
and any use of TI products in such safety-critical applications, notwithstanding any applications-related information or support that may be
provided by TI. Further, Buyers must fully indemnify TI and its representatives against any damages arising out of the use of TI products in
such safety-critical applications.
TI products are neither designed nor intended for use in military/aerospace applications or environments unless the TI products are
specifically designated by TI as military-grade or "enhanced plastic." Only products designated by TI as military-grade meet military
specifications. Buyers acknowledge and agree that any such use of TI products which TI has not designated as military-grade is solely at
the Buyer's risk, and that they are solely responsible for compliance with all legal and regulatory requirements in connection with such use.
TI products are neither designed nor intended for use in automotive applications or environments unless the specific TI products are
designated by TI as compliant with ISO/TS 16949 requirements. Buyers acknowledge and agree that, if they use any non-designated
products in automotive applications, TI will not be responsible for any failure to meet such requirements.
Following are URLs where you can obtain information on other Texas Instruments products and application solutions:
Products
Applications
Amplifiers
amplifier.ti.com
Audio
www.ti.com/audio
Data Converters
dataconverter.ti.com
Automotive
www.ti.com/automotive
DLP® Products
www.dlp.com
Communications and
Telecom
www.ti.com/communications
DSP
dsp.ti.com
Computers and
Peripherals
www.ti.com/computers
Clocks and Timers
www.ti.com/clocks
Consumer Electronics
www.ti.com/consumer-apps
Interface
interface.ti.com
Energy
www.ti.com/energy
Logic
logic.ti.com
Industrial
www.ti.com/industrial
Power Mgmt
power.ti.com
Medical
www.ti.com/medical
Microcontrollers
microcontroller.ti.com
Security
www.ti.com/security
RFID
www.ti-rfid.com
Space, Avionics &
Defense
www.ti.com/space-avionics-defense
RF/IF and ZigBee® Solutions www.ti.com/lprf
Video and Imaging
www.ti.com/video
Wireless
www.ti.com/wireless-apps
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2010, Texas Instruments Incorporated