DS250DF410ABMT_技术文档

Support &

Community

Product

Folder

Order

Now

Tools &

Software

Technical

Documents

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

DS250DF410 25Gbps 多速率 4 通道重定时器

1 特性

2 应用

1

•

具有集成信号调节功能的四通道多速率重定时器

•

针对前端口光学模块的抖动消除

有源电缆组件

所有通道均可独立锁定在 20.2752 至 25.8Gbps 的

范围内（包括 10.3125Gbps、12.5Gbps 等子速

率）

背板/中板长度延长

IEEE802.3bj 100GbE、Infiniband EDR 和 OIF-

CEI-25G-LR/MR/SR/VSR 电气接口

•

超低延迟：25.78125Gbps 数据速率下的典型延迟

< 500ps

•

SFP28、QSFP28、CFP2/CFP4、CDFP

单电源，无需低抖动参考时钟，最大限度减少电源

去耦以降低电路板布线复杂程度并节省物料清单

(BOM) 成本

3 说明

DS250DF410 是一款具有集成信号调节功能的四通道

多速率重定时器。该器件用于扩展有损且存在串扰的远

距离高速串行链路的延伸长度并提升稳定性，同时实现

不高于 10^-15的比特误码率 (BER)。

•

集成 2×2 交叉点

自适应性连续时间线性均衡器 (CTLE)

自适应判决反馈均衡器 (DFE)

带有 3 抽头有限冲激响应 (FIR) 滤波器的低抖动发

射器

DS250DF410 各通道的串行数据速率均可独立锁定在

20.2752Gbps 至 25.8Gbps 的连续范围内或者支持的

任意子速率（速率的一半和四分之一），包括

10.3125Gbps 和 12.5Gbps 等关键数据速率，因此该

器件支持独立通道前向纠错 (FEC) 直通。

•

组合式均衡，在 12.9GHz 频率下支持 35dB 以上的

通道损耗

可调节发送幅值：205mVppd 至 1225mVppd（典

型值）

片上眼图张开度监视器 (EOM)，PRBS 模式校验器

/发生器

器件信息⁽¹⁾

器件型号

封装

ABM (101)

封装尺寸（标称值）

•

支持 JTAG/AC-JTAG 边界扫描

DS250DF410

6.00mm × 6.00mm

小型 6mm × 6mm BGA 封装，可轻松实现直通布

线

(1) 如需了解所有可用封装，请参阅产品说明书末尾的可订购产品

附录。

简化原理图

RX0P

TX0P

TX0N

CDR

RX

TX

RX0N

.

RX3P

RX3N

TX3P

TX3N

2.5V or

3.3V

CDR

TX

VDD

SMBus

Slave mode

1 kΩ

To other open-drain

interrupt pins

EN_SMB

INT_N

SDA⁽¹⁾

NC_TEST

To system

SMBus

SDC⁽¹⁾

Address straps

ADDR0

(pull-up, pull-

ADDR1

down, or float)

25 MHz

To next device‘s

CAL_CLK_IN

READ_EN_N

CAL_CLK_OUT

ALL_DONE_N

SMBus Slave

mode

Float for SMBus Slave

mode, or connect to next

device‘s READ_EN_N for

SMBus Master mode

2.5V

GND

VDD

1ꢀF

(2x)

0.1ꢀF

(4x)

(1) SMBus signals need to be pulled up elsewhere in the system.

1

本文档旨在为方便起见，提供有关 TI 产品中文版本的信息，以确认产品的概要。有关适用的官方英文版本的最新信息，请访问 www.ti.com，其内容始终优先。 TI 不保证翻译的准确

性和有效性。在实际设计之前，请务必参考最新版本的英文版本。

English Data Sheet: SNLS456

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

1

2

3

4

5

6

7

特性.......................................................................... 1

应用.......................................................................... 1

说明.......................................................................... 1

修订历史记录 ........................................................... 2

说明（续）.............................................................. 3

Pin Configuration and Functions......................... 4

Specifications......................................................... 8

7.1 Absolute Maximum Ratings ...................................... 8

7.2 ESD Ratings.............................................................. 8

7.3 Recommended Operating Conditions....................... 8

7.4 Thermal Information.................................................. 9

7.5 Electrical Characteristics........................................... 9

8

9

Detailed Description ............................................ 17

8.1 Overview ................................................................. 17

8.2 Functional Block Diagram ....................................... 17

8.3 Feature Description................................................. 18

8.4 Device Functional Modes........................................ 29

8.5 Programming........................................................... 30

8.6 Register Maps......................................................... 32

Application and Implementation ........................ 76

9.1 Application Information............................................ 76

9.2 Typical Applications ............................................... 76

10 Power Supply Recommendations ..................... 90

11 Layout................................................................... 90

11.1 Layout Guidelines ................................................. 90

11.2 Layout Example .................................................... 90

12 器件和文档支持 ..................................................... 92

12.1 文档支持................................................................ 92

12.2 文档支持................................................................ 92

12.3 接收文档更新通知 ................................................. 92

12.4 支持资源................................................................ 92

12.5 商标....................................................................... 92

12.6 静电放电警告......................................................... 92

12.7 Glossary................................................................ 92

13 机械、封装和可订购信息....................................... 92

7.6 Timing Requirements, Retimer Jitter

Specifications........................................................... 13

7.7 Timing Requirements, Retimer Specifications........ 14

7.8 Timing Requirements, Recommended Calibration

Clock Specifications................................................. 14

7.9 Recommended SMBus Switching Characteristics

(Slave Mode)............................................................ 14

7.10 Recommended SMBus Switching Characteristics

(Master Mode).......................................................... 15

7.11 Recommended JTAG Switching Characteristics .. 15

7.12 Typical Characteristics ......................................... 16

4 修订历史记录

注：之前版本的页码可能与当前版本有所不同。

Changes from Revision C (February 2019) to Revision D

Page

•

首次公开发布 .......................................................................................................................................................................... 1

2

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

5 说明（续）

DS250DF410 具备一个单电源，能够最大限度地减少外部组件的数量。这些特性可降低 PCB 布线的复杂程度并

节省 BOM 成本。

DS250DF230 的高级均衡特性包括：一个低抖动 3 抽头发送有限冲激响应 (FIR) 滤波器、一个自适应连续时间线

性均衡器 (CTLE) 以及一个自适应判决反馈均衡器 (DFE)。支持针对具有多个连接器且存在串扰的有损互连和背板

进行扩展。集成的时钟和数据恢复 (CDR) 功能可重置抖动预算并对高速串行数据进行重定时，非常适用于前端口

光学模块以重置抖动容许量并重定时高速串行数据。DS250DF410 对每个通道对采用 2x2 交叉点，可为主机同时

提供通道交叉、扇出和多路复用选项。

DS250DF410 可通过 SMBus 或外部 EEPROM 进行配置。单个 EEPROM 最多可由 16 个器件使用“公共通道”格

式共享。非破坏性片上眼图监视器和 PRBS 发生器/校验器为系统内诊断提供支持。

3

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

6 Pin Configuration and Functions

ABM Package

101-Pin FC/CSP

Top View

Pin Functions

INTERNAL

PULL-UP/

PULL-DOWN

TYPE

DESCRIPTION

NAME

NO.

HIGH SPEED DIFFERENTIAL I/Os

RX0N

A10

A11

A7

Input

None

Inverting and non-inverting differential

inputs to the equalizer. An on-chip 100-Ω

termination resistor connects RXP to

RXN. These inputs need to be AC

coupled.⁽¹⁾

RX0P

RX1N

RX1P

Inverting and non-inverting differential

inputs to the equalizer. An on-chip 100-Ω

termination resistor connects RXP to

RXN. These inputs need to be AC

coupled.⁽¹⁾

A8

(1) High-speed pins do not have short-circuit protection. High-speed pins should be AC-coupled.

4

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Pin Functions (continued)

INTERNAL

PULL-UP/

PULL-DOWN

TYPE

DESCRIPTION

NAME

NO.

RX2N

A4

Input

None

Inverting and non-inverting differential

inputs to the equalizer. An on-chip 100-Ω

termination resistor connects RXP to

RXN. These inputs need to be AC

coupled.⁽¹⁾

RX2P

RX3N

RX3P

A5

A1

A2

Input

None

Inverting and non-inverting differential

inputs to the equalizer. An on-chip 100-Ω

termination resistor connects RXP to

RXN. These inputs need to be AC

coupled.⁽¹⁾

TX0N

TX0P

TX1N

TX1P

TX2N

TX2P

TX3N

TX3P

L10

L11

L7

Output

None

Inverting and non-inverting 50-Ω driver

outputs. Compatible with AC-coupled

differential inputs. These outputs need to

be AC coupled.⁽¹⁾

Inverting and non-inverting 50-Ω driver

outputs. Compatible with AC-coupled

differential inputs. These outputs need to

be AC coupled.⁽¹⁾

L8

L4

Inverting and non-inverting 50-Ω driver

outputs. Compatible with AC-coupled

differential inputs. These outputs need to

be AC coupled.⁽¹⁾

L5

L1

Inverting and non-inverting 50-Ω driver

outputs. Compatible with AC-coupled

differential inputs. These outputs need to

be AC coupled.⁽¹⁾

L2

CALIBRATION CLOCK PINS

25 MHz (±100 PPM) 2.5-V single-ended

clock from external oscillator. No stringent

phase noise or jitter requirements on this

clock. Used to calibrate VCO frequency

range.

Weak pull-up in

JTAG mode. Weak

pull-down in

CAL_CLK_IN/

JTAG_TDI

Input, 2.5V

CMOS

F1

functional mode

In JTAG mode (EN_SMB = 1 kΩ to

GND), this is JTAG Test Data In (TDI).

2.5-V buffered replica of calibration clock

input (pin E1) for connecting multiple

devices in a daisy-chained fashion.

In JTAG mode (EN_SMB = 1 kΩ to

GND), this is JTAG Test Data Out (TDO).

CAL_CLK_OUT/

JTAG_TDO

Output, 2.5V

CMOS

F11

None

SYSTEM MANAGEMENT BUS (SMBus) PINS

ADDR0

D11

Input, 4-level

4-level strap pins used to set the SMBus

address of the device. The pin state is

read on power-up. The multi-level nature

of these pins allows for 16 unique device

addresses. The four strap options include:

0: 1 kΩ to GND

R: 20 kΩ to GND

F: Float

1: 1 kΩ to VDD

Weak pull-up in

JTAG mode

ADDR1/JTAG_TRS

D1

Refer to Device SMBus Address for more

information.

In JTAG mode (EN_SMB = 1 kΩ to

GND), ADDR1 is JTAG Test Reset

(TRS).

5

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Pin Functions (continued)

INTERNAL

TYPE

PULL-UP/

PULL-DOWN

DESCRIPTION

NAME

NO.

Four-level 2.5-V input used to select

between SMBus master mode (float) and

SMBus slave mode (high). The four

defined levels are:

0: 1 kΩ to GND - JTAG mode; certain

pins take on JTAG functionality

R: 20 kΩ to GND - RESERVED, TI test

mode

EN_SMB

G1

Input, 4-level

None

F: Float - SMBus Master Mode

1: 1 kΩ to VDD - SMBus Slave Mode

SMBus data input / open drain output.

External 2-kΩ to 5-kΩ pull-up resistor is

required as per SMBus interface

I/O, 3.3V

LVCMOS, Open

Drain

SDA

G11

H11

None

standard. This pin is 3.3V-tolerant.

SMBus clock input / open drain clock

output. External 2-kΩ to 5-kΩ pull-up

resistor is required as per SMBus

interface standard. This pin is 3.3V-

tolerant.

I/O, 3.3V

LVCMOS, Open

Drain

SDC

SMBus MASTER MODE PINS

Indicates the completion of a valid

EEPROM register load operation when in

SMBus Master Mode (EN_SMB=Float):

High = External EEPROM load failed or

incomplete

Output,

LVCMOS

ALL_DONE_N

E1

None

Low = External EEPROM load successful

and complete

When in SMBus slave mode

(EN_SMB=1), this output reflects the

status of the READ_EN_N input.

SMBus Master Mode (EN_SMB=Float):

When asserted low, initiates the SMBus

master mode EEPROM read function.

Once EEPROM read is complete

(indicated by assertion of ALL_DONE_N

low), this pin can be held low for normal

device operation.

SMBus Slave Mode (EN_SMB=1): When

asserted low, this causes the device to be

held in reset (SMBus state machine reset

and register reset). This pin should be

pulled high or left floating for normal

operation in SMBus Slave Mode.

This pin is 3.3V-tolerant.

Input, 3.3V

LVCMOS

READ_EN_N

E11

Weak pull-up

MISCELLANEOUS PINS

Open-drain 3.3-V tolerant active-low

interrupt output. This pin is pulled low

when an interrupt occurs. The events

which trigger an interrupt are

programmable through SMBus registers.

INT_N can be connected in a wired-OR

fashion with other device's interrupt pin. A

single pull-up resistor in the 2-kΩ to 5-kΩ

range is adequate for the entire INT_N

net.

Output,

LVCMOS, Open

Drain

INT_N

H1

None

6

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Pin Functions (continued)

INTERNAL

PULL-UP/

PULL-DOWN

TYPE

DESCRIPTION

NAME

NO.

No pull-up in JTAG Reserved TI test pins. During normal

TEST0/JTAG_TCK

D10

Input, LVCMOS mode. Weak pull-up (non-test-mode) operation, these pins are

in functional mode. configured as inputs and therefore they

are not affected by the presence of a

signal. These pins may be left floating,

tied to GND, or connected to a 2.5-V

(max) output.

TEST1/JTAG_TMS

D2

Input, LVCMOS

Weak pull-up

In JTAG mode (EN_SMB = 1 kΩ to

GND), TEST0 is JTAG Test Clock (TCK)

and TEST1 is JTAG Test Mode Select

(TMS).

TEST2

TEST3

TEST4

E10

E2

Input, LVCMOS

Weak pull-up

Reserved TI test pins. During normal

(non-test-mode) operation, these pins are

configured as inputs and therefore they

are not affected by the presence of a

signal. These pins may be left floating,

tied to GND, or connected to a 2.5-V

(max) output.

H10

TEST5

H2

Input, LVCMOS

Weak pull-up

POWER

A3, A6, A9, B1, B2, B3,

B4, B5, B6, B7, B8, B9,

B10, B11, C1, C6, C11,

D6, E3, E5, E7, E9, F3,

F4, F5, F6, F7, F8, F9,

Ground reference. The GND pins on this

device should be connected through a

low-resistance path to the board GND

plane.

GND

Power

None

G2, G3, G5, G7, G9, G10,

H6, J1, J6, J11, K1, K2,

K3, K4, K5, K6, K7, K8,

K9, K10, K11, L3, L6, L9

Power supply, VDD = 2.5 V ±5%. TI

recommends connecting at least six de-

coupling capacitors between the

DS250DF410’s VDD plane and GND as

close to the DS250DF410 as possible.

For example, four 0.1-μF capacitors and

two 1-μF capacitors directly beneath the

device or as close to the VDD pins as

possible. The VDD pins on this device

should be connected through a low-

resistance path to the board VDD plane.

C3, C9, D3, D4, D5, D7,

D8, D9, H3, H4, H5, H7,

H8, H9, J3, J9

VDD

Power

None

7

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

7 Specifications

7.1 Absolute Maximum Ratings

over operating free-air temperature range (unless otherwise noted)

(1)

MIN

–0.5

MAX

UNIT

V

VDD_ABSMAX

VIO_2.5V,ABSMAX

VIO_3.3V,ABSMAX

VIN_ABSMAX

VOUT_ABSMAX

TJ_ABSMAX

Supply voltage (VDD)

2.75

4.0

2.5 V I/O voltage (LVCMOS, CMOS and Analog)

V

Open Drain Voltage (SDA, SDC, INT_N) and LVCMOS Input Voltage (READ_EN_N)

Signal input voltage (RXnP, RXnN)

V

2.75

150

V

Signal output voltage (TXnP, TXnN)

Junction temperature

V

ºC

T_stg

Storage temperature

–40

(1) Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. These are stress ratings

only, which do not imply functional operation of the device at these or any other conditions beyond those indicated under Recommended

Operating Conditions. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability.

7.2 ESD Ratings

VALUE

UNIT

Human-body model (HBM), per ANSI/ESDA/JEDEC JS-001⁽¹⁾

±2000

V_(ESD)

Electrostatic discharge

V

Charged-device model (CDM), per JEDEC specification JESD22-

C101⁽²⁾

±1000

(1) JEDEC document JEP155 states that 500-V HBM allows safe manufacturing with a standard ESD control process. Pins listed as ±2 kV

may actually have higher performance.

(2) JEDEC document JEP157 states that 250-V CDM allows safe manufacturing with a standard ESD control process.

7.3 Recommended Operating Conditions

over operating free-air temperature range (unless otherwise noted)

MIN

MAX UNIT

2.625

VDD

Supply voltage, VDD to GND. DC plus AC power should not exceed these limits.

Supply noise, DC to < 50 Hz, sinusoidal⁽¹⁾

Supply noise, 50 Hz to 10 MHz, sinusoidal⁽¹⁾

Supply noise, >10 MHz, sinusoidal⁽¹⁾

2.375

V

NVDD

T_rampVDD

T_J

250 mVpp

20 mVpp

10 mVpp

μs

VDD supply ramp time, from 0V to 2.375V

Operating junction temperature

150

-40

110

85⁽²⁾

2.625

3.6

ºC

V

T_A

Operating ambient temperature

-40

VIO_2.5V

VIO_{3.3V,INT_N}

VIO_3.3V

2.5 V I/O voltage (LVCMOS, CMOS and Analog)

Open Drain LVCMOS I/O voltage (INT_N)

Open Drain LVCMOS I/O voltage (SDA, SDC)

2.375

V

2.375

3.6

V

(1) Steps must be taken to ensure the combined AC plus DC supply noise meets the specified VDD supply voltage limits.

(2) Steps must be taken to ensure the operating junction temperature range and ambient temperature stay-in-lock range (TEMP_LOCK+

TEMP_LOCK-) are met. Refer to Electrical Characteristics for more details concerning TEMP_LOCK+and TEMP_LOCK-

,

.

8

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

7.4 Thermal Information

DS250DF410

THERMAL METRIC⁽¹⁾

CONDITIONS/ASSUMPTIONS⁽²⁾

FC/CSP (ABM)

101 PINS

34.2

UNIT

4-Layer JEDEC Board

10-Layer 8-in x 6-in Board

20-Layer 8-in x 6-in Board

30-Layer 8-in x 6-in Board

4-Layer JEDEC Board

16.7

R_θJA

Junction-to-ambient thermal resistance

°C/W

15.6

14.6

R_θJC(top)

R_θJB

Junction-to-case (top) thermal resistance

Junction-to-board thermal resistance

7.8

°C/W

4-Layer JEDEC Board

13.7

4-Layer JEDEC Board

0.004

13.0

10-Layer 8-in x 6-in Board

20-Layer 8-in x 6-in Board

30-Layer 8-in x 6-in Board

4-Layer JEDEC Board

Ψ_JT

Junction-to-top characterization parameter

Junction-to-board characterization parameter

°C/W

10-Layer 8-in x 6-in Board

20-Layer 8-in x 6-in Board

30-Layer 8-in x 6-in Board

11.7

Ψ_JB

11.5

11.3

(1) For more information about traditional and new thermal metrics, see the Semiconductor and IC Package Thermal Metrics application

report, SPRA953.

(2) No heat sink or airflow was assumed for these estimations. Depending on the application, a heat sink, faster airflow, and/or reduced

ambient temperature (<85 C) may be required in order to meet the maximum junction temperature specification per the Recommended

Operating Conditions.

7.5 Electrical Characteristics

over operating free-air temperature range (unless otherwise noted)

PARAMETER

TEST CONDITIONS

Full-rate

MIN

20.2752

10.1376

5.0688

TYP

MAX

25.8

12.9

6.45

UNIT

Gbps

Rbaud

Input data rate

Half-rate

Quarter-rate

Single device reading its

configuration from an EEPROM.

Common channel configuration. This

time scales with the number of

devices reading from the same

EEPROM.

t_EEPROM

EEPROM configuration load time

Power-on reset assertion time

15⁽¹⁾

40⁽¹⁾

50

ms

Single device reading its

configuration from an EEPROM.

Unique channel configuration. This

time scales with the number of

devices reading from the same

EEPROM.

t_EEPROM

Internal power-on reset (PoR)

stretch between stable power supply

and de-assertion of internal PoR.

The SMBus address is latched on

the completion of the PoR stretch,

and SMBus accesses are permitted.

t_POR

(1) From low assertion of READ_EN_N to low assertion of ALL_DONE_N. Does not include Power-On Reset time.

9

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Electrical Characteristics (continued)

over operating free-air temperature range (unless otherwise noted)

PARAMETER

POWER SUPPLY

TEST CONDITIONS

MIN

TYP

241

233

MAX

UNIT

mW

With CTLE, full DFE, Tx FIR,

Driver, and Crosspoint enabled. Idle

power consumption not included.

305

With CTLE, full DFE, Tx FIR, and

Driver enabled; Crosspoint disabled.

Idle power consumption not

included.

mW

With CTLE, partial DFE (taps 1-2

only), Tx FIR, and Driver enabled;

Crosspoint and DFE taps 3-5

disabled. Idle power consumption

not included.

220

mW

With CTLE, Tx FIR, Driver, and

Crosspoint enabled; DFE disabled.

Idle power consumption not

included.

211

365

290

430

mW

Power consumption per active

channel

W_channel

Assuming CDR acquiring lock with

CTLE, full DFE, Tx FIR, Driver, and

Crosspoint enabled. Idle power

consumption not included.

Assuming CDR acquiring lock with

CTLE, Tx FIR, Driver, and

Crosspoint enabled; DFE disabled.

Idle power consumption not

included.

318

393

mW

PRBS checker power consumption

only⁽²⁾

220

230

302

315

mW

PRBS generator power power

consumption only⁽²⁾

Total idle power consumption

Idle/static mode, power supplied, no

high-speed data present at inputs,

all channels automatically powered

down.

W_{static_total}

329

525

665

mW

With CTLE, full DFE, Tx FIR, Driver,

and Crosspoint enabled.

518

505

mA

With CTLE, full DFE, Tx FIR, and

Driver enabled; Crosspoint disabled.

Active mode total device supply

current consumption

With CTLE, partial DFE (taps 1-2

only), Tx FIR, and Driver enabled;

Crosspoint and DFE taps 3-5

disabled.

I_total

485

470

132

mA

With CTLE, Tx FIR, Driver, and

Crosspoint enabled. DFE disabled.

639

200

Idle/static mode. Power supplied, no

Idle mode total device supply current high-speed data present at inputs,

consumption

I_{static_total}

all channels automatically powered

down.

LVCMOS DC SPECIFICATIONS

2.5-V LVCMOS pins

1.75

VDD

3.6

V

V_IH

V_IL

Input high-level voltage

Input low-level voltage

3.3-V LVCMOS pin (READ_EN_N)

2.5-V LVCMOS pins

GND

0.7

3.3-V LVCMOS pin (READ_EN_N)

0.8

(2) To ensure optimal performance, it is recommended to not enable more than two PRBS blocks (checker and/or generator) per device.

10

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Electrical Characteristics (continued)

over operating free-air temperature range (unless otherwise noted)

PARAMETER

TEST CONDITIONS

MIN

TYP

MAX

UNIT

High level (1) input voltage

4-level pins ADDR0, ADDR1, and

EN_SMB

0.95 ×

VDD

V

Float level input voltage

10K to GND input voltage

Low -evel (0) input voltage

4-level pins ADDR0, ADDR1, and

EN_SMB

0.67 ×

VDD

V

V_TH

4-level pins ADDR0, ADDR1, and

EN_SMB

0.33 ×

VDD

4-level pins ADDR0, ADDR1, and

EN_SMB

0.1

V_OH

V_OL

I_IH

High-level output voltage

Low-level output voltage

Input high leakage current

IOH = 4 mA

2

V

IOL = –4 mA

0.4

70

65

Vinput = VDD, Open drain pins

Vinput = VDD and CAL_CLK_IN pin

μA

I_IH

Vinput = VDD, ADDR[1:0] and

EN_SMB pins

I_IH

Input high leakage current

65

15

μA

I_IH

I_IL

Input high leakage current

Input low leakage current

Vinput = VDD, READ_EN_N

Vinput = 0 V, Open drain pins

Vinput = 0 V, CAL_CLK_IN pins

μA

-15

Vinput = 0 V, ADDR[1:0],

READ_EN_N, and EN_SMB pins

I_IL

Input low leakage current

-115

μA

RECEIVER INPUTS (RXnP, RXnN)

V_IDMax

Maximum input differential voltage

For normal operation

1225

<-18

<-14

mVppd

dB

Between 50 MHz and 3.69 GHz

Between 3.69 GHz and 12.9 GHz

Differential input return loss,

SDD11⁽³⁾

RL_SDD11

dB

Differential to common-mode input

return loss, SDC11⁽³⁾

RL_SDC11

RL_SCD11

RL_SCC11

Between 50 MHz and 12.9 GHz

<-23

<-22

dB

Differential to common-mode input

return loss, SCD11⁽³⁾

Between 150 MHz and 10 GHz

Between 10 GHz and 12.9 GHz

<-11

<-10

dB

Common-mode input return loss,

SCC11⁽³⁾

Minimum input peak-to-peak

amplitude level at device pins

required to assert signal detect.

25.78125Gbps with PRBS7 pattern

and 20dB loss channel

AC signal detect assert (ON)

threshold level

V_SDAT

196

147

mVppd

Maximum input peak-to-peak

amplitude level at device pins which

causes signal detect to de-assert.

25.78125Gbps with PRBS7 pattern

and 20dB loss channel

AC signal detect de-assert (OFF)

threshold level

V_SDDT

TRANSMITTER OUTPUTS (TXnP, TXnN)

Measured with c(0)=7 setting

(Reg_0x3D[6:0]=0x07,

Reg_0x3E[6:0]=0x40,

REG_0x3F[6:0]=0x40). Differential

VOD

Output differential voltage amplitude measurement using an 8T pattern

(eight 1s followed by eight 0s) at

25.78125 Gbps with TXPn and

525

mVppd

TXNn terminated by 50 Ohms to

GND.

(3) Measured with an evaluation board which uses microstrip traces and low-loss dielectric with approximately 3 dB insertion loss at 12.9

GHz between the DS250DF410 and the measurement instrument.

11

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Electrical Characteristics (continued)

over operating free-air temperature range (unless otherwise noted)

PARAMETER

TEST CONDITIONS

MIN

TYP

MAX

UNIT

Measured with c(0)=31 setting

(Reg_0x3D[6:0]=0x1F,

Reg_0x3E[6:0]=0x40,

REG_0x3F[6:0]=0x40). Differential

VOD

Output differential voltage amplitude measurement using an 8T pattern

(eight 1s followed by eight 0s) at

25.78125 Gbps with TXPn and

1225

mVppd

TXNn terminated by 50 Ohms to

GND.

Differential output amplitude with TX

disabled

VOD_idle

VOD_res

< 11

< 50

mVppd

Difference in VOD between two

adjacent c(0) settings. Applies to

VOD in the 525mVppd to

Output VOD resolution

1225mVppd range [c(0)>4].

With respect to signal ground.

Measured with PRBS9 data pattern.

Common-mode AC output noise

V_cm-TX-AC

6.5

17

mV, RMS

Measured with a 33GHz (-3dB) low-

pass filter.

20%-to-80% rise time and 80%-to-

20% fall time on a clock-like {11111

00000} data pattern at 25.78125

t_r, t_f

Output transition time

Gbps. Measured for ~800 mVppd

output amplitude and no

ps

equalization: Reg_0x3D=+13,

Reg_0x3E=0, REG_0x3F=0

Between 50 MHz and 5 GHz

Between 5 GHz and 12.9 GHz

<-14

<-13

dB

Differential output return loss,

SDD22⁽⁴⁾

RL_SDD22

Common-mode to differential output

return loss, SCD22⁽⁴⁾

RL_SCD22

RL_SDC22

Between 50 MHz and 12.9 GHz

<-22

dB

Differential-to-common-mode output

return loss, SDC22⁽⁴⁾

Between 50 MHz and 10 GHz

Between 10 GHz and 12.9 GHz

<-9

dB

Common-mode output return loss,

SCC22⁽⁴⁾

RL_SCC22

<-12

SMBus ELECTRICAL CHARACTERISTICS (SLAVE MODE)

V_IH

V_IL

Input high-level voltage

Input low-level voltage

Input pin capacitance

Low-level output voltage

SDA and SDC

1.75

3.6

0.8

V

GND

C_IN

V_OL

15

pF

V

SDA or SDC, IOL = 1.25 mA

0.4

15

SDA or SDC, VINPUT = VIN, VDD,

GND

I_IN

Input current

-15

μA

T_R

T_F

SDA rise time, read operation

SDA fall time, read operation

Pull-up resistor = 1 kΩ, Cb = 50pF

150

4.5

ns

(4) Measured with an evaluation board which uses microstrip traces and low-loss dielectric with approximately 3 dB insertion loss at 12.9

GHz between the DS250DF410 and the measurement instrument.

12

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

7.6 Timing Requirements, Retimer Jitter Specifications

over operating free-air temperature range (unless otherwise noted)

PARAMETER

TEST CONDITIONS

MIN

TYP

MAX

UNIT

Measured at 25.78125 Gbps to a

probability level of 1E-12 with

PRBS11 data pattern an evaluation

board traces de-embedded.

UIpp @

1E-12

J_TJ

Output total jitter (TJ)

0.17

Measured at 25.78125 Gbps to a

probability level of 1E-12 with

PRBS11 data pattern an evaluation

board traces de-embedded

J_RJ

Output random Jitter (RJ)

Output duty cycle distortion (DCD)

Jitter peaking

6

4

mUI RMS

mUIpp

dB

Measured at 25.78125 Gbps to a

probability level of 1E-12 with

PRBS11 data pattern an evaluation

board traces de-embedded

J_DCD

J_PEAK

Measured at 10.3125 Gbps with

PRBS7 data pattern. Peaking

frequency in the range of 1 to 6

MHz.

0.8

0.4

Measured at 25.78125 Gbps with

PRBS7 data pattern. Peaking

frequency in the range of 1 to 17

MHz.

Jitter peaking

dB

Data rate of 10.3125Gbps with

PRBS7 pattern

BWPLL

PLL bandwidth

5.3

5.5

MHz

Data rate of 25.78125Gbps with

PRBS7 pattern

Measured at 25.78125 Gbps with

SJ frequency = 190 KHz, 30dB

input channel loss, PRBS31 data

pattern, 800 mVppd launch

amplitude, and 0.078 UIpp total

uncorrelated output jitter in addition

to the applied SJ. BER < 1E-12.

J_TOL

Input jitter tolerance

9

1

UIpp

Measured at 25.78125 Gbps with

SJ frequency = 940 KHz, 30dB

input channel loss, PRBS31 data

pattern, 800 mVppd launch

amplitude, and 0.078 UIpp total

uncorrelated output jitter in addition

to the applied SJ. BER < 1E-12.

Measured at 25.78125 Gbps with

SJ frequency > 15MHz, 30dB input

channel loss, PRBS31 data pattern,

800 mVppd launch amplitude, and

0.078 UIpp total uncorrelated output

jitter in addition to the applied SJ.

BER < 1E-12.

0.3

CDR stay-in-lock ambient

temperature range, negative ramp.

85 °C starting ambient temperature,

ramp rate -3 °C/minute, 1.7

liters/sec airflow, 12 layer PCB.

TEMP_LOCK-Maximum temperature change

below initial CDR lock acquisition

temperature.

115

125

°C

CDR stay-in-lock ambient

temperature range, positive ramp.

TEMP_LOCK+Maximum temperature change

above initial CDR lock acquisition

temperature.

-40 °C starting ambient

temperature, ramp rate +3

°C/minute, 1.7 liters/sec airflow, 12

layer PCB.

13

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

7.7 Timing Requirements, Retimer Specifications

over operating free-air temperature range (unless otherwise noted)

PARAMETER

TEST CONDITIONS

MIN

TYP

MAX

UNIT

No crosspoint; CDR enabled and

locked.

3.5UI +

125ps

ps

Input-to-output latency (propagation Crosspoint enabled; CDR enabled

3.5UI +

145ps

t_D

ps

delay) through a channel

and locked.

No crosspoint; CDR in raw (bypass)

mode.

< 145

Latency difference between

channels at full-rate. 25.78125 Gbps

data rate.

t_SK

Channel-to-channel interpair skew

CDR lock acquisition time

< 30

ps

Measured at 25.78125 Gbps, Adapt

Mode = 1 (Reg_0x31[6:5]=0x1),

EOM timer = 0x5

t_lock

< 100

ms

(Reg_0x2A[7:4]=0x5).

Measured at 10.3125 Gbps, Adapt

Mode = 1 (Reg_0x31[6:5]=0x1),

EOM timer = 0x5

t_lock

CDR lock acquisition time

ms

(Reg_0x2A[7:4]=0x5).

7.8 Timing Requirements, Recommended Calibration Clock Specifications

over operating free-air temperature range (unless otherwise noted)

PARAMETER

TEST CONDITIONS

MIN

TYP

MAX

UNIT

MHz

PPM

CLK_f

Calibration clock frequency

Calibration clock PPM tolerance

25

CLK_PPM

-100

40%

100

Recommended/tolerable input duty

cycle

CLK_IDC

50%

60%

Intrinsic duty cycle distortion of chip

calibration clock output at the

CAL_CLK_OUT pin, assuming 50%

duty cycle on CAL_CLK_IN pin.

Intrinsic calibration clock duty cycle

distortion

CLK_ODC

45%

55%

Assumes worst-case 60%/40% input

duty cycle on the first device.

CAL_CLK_OUT from first devuce

connects to CAL_CLK_IN of second

device, and so on until the last

device.

Number of devices which can be

cascaded from CAL_CLK_OUT to

CAL_CLK_IN

CLKnum

20

N/A

7.9 Recommended SMBus Switching Characteristics (Slave Mode)

over operating free-air temperature range (unless otherwise noted)

PARAMETER

SDC clock frequency

Data hold time

TEST CONDITIONS

MIN

TYP

100

0.75

100

MAX

UNIT

kHz

ns

f_SDC

10

400

t_HD-DAT

t_SU-DAT

Data setup time

ns

14

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

7.10 Recommended SMBus Switching Characteristics (Master Mode)

over operating free-air temperature range (unless otherwise noted)

PARAMETER

SDC clock frequency

SDC low period

TEST CONDITIONS

MIN

260

TYP

303

1.90

1.40

0.6

MAX

346

UNIT

kHz

μs

f_SDC

T_LOW

1.66

1.22

2.21

1.63

T_HIGH

T_HD-STA

T_SU-STA

T_HD-DAT

T_SD-DAT

T_SU-STO

T_BUF

SDC high period

μs

Hold time start operation

Setup time start operation

Data hold time

μs

0.6

μs

0.6

μs

Data setup time

0.1

μs

Stop condition setup time

Bus free time between Stop-Start

SDC rise time

0.6

μs

1.3

μs

T_R

Pull-up resistor = 1 kΩ

Pull-up resistor = 1 kΩ

300

ns

T_F

SDC fall time

ns

7.11 Recommended JTAG Switching Characteristics

over operating free-air temperature range (unless otherwise noted)

PARAMETER

TEST CONDITIONS

MIN

TYP

MAX

UNIT

ns

t_TCK

t_SU

TCK clock period

100

TDI, TMS setup time to TCK

TDI, TMS hold time to TCK

TCK falling edge to TDO

50

ns

t_HD

ns

t_DLY

ns

15

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

7.12 Typical Characteristics

1.6

1.4

1.2

1

c(0)=7

c(0)=16

c(0)=31

1.4

c(0)=31

1.2

1

0.8

0.6

0.4

0.8

0.6

0.4

2.325

2.395

2.465

2.535

2.605

2.675

-40

-15

10

35

60

85

VDD Supply Voltage (V)

Ambient Temperature (°C)

C001

C002

图 1. Typical VOD vs Supply Voltage

图 2. Typical VOD vs Temperature

0.25

0.2

0.15

0.1

0.05

0

1.4

1.3

1.2

1.1

1

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

T = 25C, VDD = 2.35V

T = 25C, VDD = 2.65V

T = 95C, VDD = 2.35V

T = 95C, VDD = 2.65V

T = -40C, VDD = 2.35V

T = -40C, VDD = 2.65V

TJ, VDD = 2.35V

TJ, VDD = 2.65V

DJ, VDD = 2.35V

DJ, VDD = 2.65V

0

5

10

15

20

25

30

-40

-30

-20

-10

0

10

20

30

40

50

60

70

80

90

Main-Cursor

Temperature (degrees C)

C002

C001

图 3. Typical VOD vs FIR Main-Cursor

图 4. Typical Output Jitter vs Temperature

at 25.78125 Gbps

45

40

35

30

25

20

15

10

5

3

2.8

2.6

2.4

2.2

2

VDD = 2.35V

VDD = 2.5V

VDD = 2.65V

VDD = 2.35V

VDD = 2.5V

VDD = 2.65V

1.8

1.6

1.4

1.2

1

0.8

0.6

0.4

0.2

0

0.1

1

10

100

0.5

1

2

4

8

16

32

64

Frequency (MHz)

C003

C004

0.1 MHz to 100 MHz

T = 25°C

Input Random Jitter = 0.078 UIpp

0.5 MHz to 100 MHz

T = 25°C

Input Random Jitter = 0.078 UIpp

图 5. Typical Sinusoidal Input Jitter Tolerance for

图 6. Typical Input Jitter Tolerance for

30-dB Channel at 25.78125 Gbps

16

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

8 Detailed Description

8.1 Overview

The DS250DF410 is a four-channel multi-rate retimer with integrated signal conditioning. Each of the four

channels operates independently. Each channel includes a continuous-time linear equalizer (CTLE) and a

Decision Feedback Equalizer (DFE), which together compensate for the presence of a dispersive transmission

channel between the source transmitter and the DS250DF410 receiver. The CTLE and DFE are self-adaptive.

Each channel includes an independent voltage-controlled oscillator (VCO) and phase-locked loop (PLL) which

produce a clean clock that is frequency-locked to the clock embedded in the input data stream. The high-

frequency jitter on the incoming data is attenuated by the PLL, producing a clean clock with substantially-reduced

jitter. This clean clock is used to re-time the incoming data, removing high-frequency jitter from the data stream

and reproducing the data on the output with significantly-reduced jitter.

Each channel of the DS250DF410 features an output driver with adjustable differential output voltage and output

equalization in the form of a three-tap finite impulse response (FIR) filter. The output FIR compensates for

dispersion in the transmission channel at the output of the DS250DF410.

Between each group of two adjacent channels (e.g. between channels 0 and 1, 2 and 3) is a full 2x2 cross-point

switch. This allows multiplexing and de-multiplexing/fanout applications for fail-over redundancy, as well as

cross-over applications to aid PCB routing.

Each channel also includes diagnostic features such as a Pseudo Random Bit Sequence (PRBS) pattern

generator and checker, as well as a non-destructive eye opening monitor (EOM). The EOM can be used to plot

the post-equalized eye at the input to the decision slicer or simply to read the horizontal eye opening (HEO) and

vertical eye opening (VEO).

The DS250DF410 is configurable through a single SMBus port. The DS250DF410 can also act as an SMBus

master to configure itself from an EEPROM. Up to sixteen DS250DF410 devices can share a single SMBus.

The sections which follow describe the functionality of various circuits and features within the DS250DF410. For

more information about how to program or operate these features, consult the DS250DF410 Programming

Guide.

8.2 Functional Block Diagram

One of Four Channels

To adjacent

channel

DFE

Term

Raw

Retimed

PRBS

RXnP

RXnN

TXnP

TXnN

CTLE+VGA

X-point

+

Sampler

TX FIR

Driver

PRBS

Gen

Signal

Detect

Voltage

Regulator

Eye

Monitor

PRBS

Checker

Voltage

Regulator

PFD, CDR,

and Divider

VCO

Channel Digital Core

CAL_CLK_IN

ADDRn

Buffer

CAL_CLK_OUT

SCL

SDA

Power-On

Reset

Always-On

10 MHz

Shared Digital Core

READ_EN_N

EN_SMB

ALL_DONE_N

INT_N

Shared Digital Core

17

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

8.3 Feature Description

8.3.1 Device Data Path Operation

The DS250DF410 data path consists of several key blocks as shown in the functional block diagram. These key

circuits are:

•

Signal Detect

Continuous Time Linear Equalizer (CTLE)

Variable Gain Amplifier (VGA)

Cross-Point Switch

Decision Feedback Equalizer (DFE)

Clock and Data Recovery (CDR)

Calibration Clock

Differential Driver with FIR Filter

8.3.2 Signal Detect

The DS250DF410 receiver contains a signal detect circuit. The signal detect circuit monitors the energy level on

the receiver inputs and powers on or off the rest of the high-speed data path if a signal is detected or not. By

default, each channel allows the signal detect circuit to automatically power on or off the rest of the high speed

data path depending on the presence of an input signal. The signal detect block can be manually controlled in

the SMBus channel registers. This can be useful if it is desired to manually force channels to be disabled. For

information on how to manually operate the signal detect circuit refer to the DS250DF410 Programming Guide.

8.3.3 Continuous Time Linear Equalizer (CTLE)

The CTLE in the DS250DF410 is a fully-adaptive equalizer. The CTLE adapts according to a Figure of Merit

(FOM) calculation during the lock acquisition process. The FOM calculation is based upon the horizontal eye

opening (HEO) and vertical eye opening (VEO). Once the CDR locks and the CTLE adapts, the CTLE boost

level is frozen until a manual re-adapt command is issued or until the CDR re-enters the lock acquisition state.

The CTLE can be re-adapted by resetting the CDR.

The CTLE consists of 4 stages, with each stage having 2-bit boost control. This allows for 256 different boost

combinations. The CTLE adaption algorithm allows the CTLE to adapt through 16 of these boost combinations.

These 16 boost combinations comprise the EQ Table in the channel registers. See channel registers 0x40

through 0x4F. This EQ Table can be reprogrammed to support up to 16 of the 256 boost settings.

The boost levels can be set between 8 dB and 25 dB (at 12.89 GHz.)

8.3.4 Variable Gain Amplifier (VGA)

The DS250DF410 receiver implements a VGA. The VGA assists in the recovery of extremely small signals,

working in conjunction with the CTLE to equalize and scale amplitude. The VGA has 1-bit control via

Reg_0x8E[0], and the VGA is in the low-gain state (Reg_0x8E[0]=0) by default. In addition to the VGA, the CTLE

implements its own gain control via Reg_0x13[5] to adjust the DC amplitude similar to the VGA. For more

information on how to configure the VGA and EQ gain, refer to the DS250DF410 Programming Guide.

8.3.5 Cross-Point Switch

Each group of two adjacent channels in the DS250DF410 has a 2×2 cross-point that may be enabled to

implement a 2-to-1 mux, a 1-to-2 fanout, or an A-to-B/B-to-A lane cross. A cross-point exists between the

following channel pairs:

•

Channel 0 and Channel 1

Channel 2 and Channel 3

18

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Feature Description (接下页)

8.3.6 Decision Feedback Equalizer (DFE)

A 5-tap DFE can be enabled within the data path of each channel to assist with reducing the effects of crosstalk,

reflections, or post cursor inter-symbol interference (ISI). The DFE must be manually enabled, regardless of the

selected adapt mode. Once the DFE has been enabled it can be configured to adapt only during lock acquisition

or to adapt continuously. The DFE can also be manually configured to specified tap polarities and tap weights.

However, when the DFE is configured manually the DFE auto-adaption should be disabled. For many

applications with lower insertion loss (i.e. < 30 dB) lower crosstalk, and/or lower reflections, part or all of the DFE

can be disabled to reduce power consumption. The DFE can either be fully enabled (taps 1-5), partially enabled

(taps 1-2 only), or fully disabled (no taps).

The DFE taps are all feedback taps with 1UI spacing. Each tap has a specified boost weight range and polarity

bit.

表 1. DFE Tap Weights

DFE PARAMETER

Tap 1 Weight Range

Tap 2-5 Weight Range

Tap Weight Step Size

DECIMAL (REGISTER VALUE)

VALUE (mV) (TYP)

0 - 31

0 - 15

NA

0 – 217

0 – 105

7

0: (+) positive; feedback value creates a low-pass filter response, thus providing attenuation to

correct for negative-sign post-cursor ISI

Polarity

1: (-) negative; Feedback value creates a high-pass filter response, thus providing boost to correct

for positive-sign post-cursor ISI.

8.3.7 Clock and Data Recovery (CDR)

The CDR consists of a Phase Locked Loop (PLL), PPM counter, and Input and Output Data Multiplexers (mux)

allowing for retimed data, un-retimed data, PRBS generator and output muted modes.

By default, the equalized data is fed into the CDR for clock and data recovery. The recovered data is then output

to the FIR filter and differential driver together with the recovered clock which has been cleaned of any high-

frequency jitter outside the bandwidth of the CDR clock recovery loop. The bandwidth of the CDR defaults to 5.5

MHz (typ) in full-rate (divide-by-1) mode and 5.3 MHz (typ) in sub-rate mode. The CDR bandwidth is adjustable.

Refer to the DS250DF410 Programming Guide for more information on adjusting the CDR bandwidth. Users can

configure the CDR data to route the recovered clock and data to the PRBS checker. Users also have the option

of configuring the output of the CDR to send raw non-retimed data, or data from the pattern generator.

The CDR requires the following to be properly configured:

•

25 MHz calibration clock to run the PPM counter (CAL_CLK_IN).

Expected data rates must be programmed into the CDR either through the rate table or entered manually with

the corrected divider settings. Refer to the Programming Guide for more information on configuring the CDR

for different data rates.

8.3.8 Calibration Clock

The calibration clock is not part of the CDR’s PLL and thus is not used for clock and data recovery. The

calibration clock is connected only to the PPM counter for each CDR. The PPM counter constrains the allowable

lock ranges of the CDR according to the programmed values in the rate table or the manually entered data rates.

The host should provide an input calibration clock signal of 25 MHz frequency. Because this clock is not used for

clock and data recovery, there are no stringent jitter requirements placed on this 25 MHz calibration clock.

8.3.9 Differential Driver with FIR Filter

The DS250DF410 output driver has a three-tap finite impulse response (FIR) filter which allows for pre- and post-

cursor equalization to compensate for a wide variety of output channel media. The filter consists of a weighted

sum of three consecutive retimed bits as shown in the following diagram. C[0] can take on values in the range [-

31, +31]. C[-1] and C[+1] can take on values in the range [-15, 15].

19

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Retimed

Data

FIR filter

output

x

+

1 UI

Delay

C[-1]

Pre-cursor

x

1 UI

Delay

C[0]

Main-curosr

x

C[+1]

Post-cursor

图 7. FIR Filter Functional model

When utilizing the FIR filter, it is important to abide by the following general rules:

•

|C[-1]| + |C[0]| + |C[+1]| ≤ 31; the FIR tap coefficients absolute sum must be less or equal to 31

sgn(C[-1]) = sgn(C[+1]) ≠ sgn(C[0]), for high-pass filter effect; the sign for the pre-cursor and/or post-cursor

tap must be different from main-cursor tap to realize boost effect

•

sgn(C[-1]) = sgn(C[+1]) = sgn(C[0]), for low-pass filter effect; the sign for the pre-cursor and/or post-cursor tap

must be equal to the main-cursor tap to realize attenuation effect

The FIR filter is used to pre-distort the transmitted waveform to compensate for frequency-dependant loss in the

output channel. The most common way of pre-distorting the signal is to accentuate the transitions and de-

emphasize the non-transitions. The bit before a transition is accentuated via the pre-cursor tap, and the bit after

the transition is accentuated via the post-cursor tap. The figures below give a conceptual illustration of how the

FIR filter affects the output waveform. The following characteristics can be derived from the example waveforms.

•

VOD_pk-pk= v₇- v₈

VOD_{low-frequency}= v₂- v₅

Rpre_dB= 20 * log₁₀(v₃⁄v₂)

Rpst_dB= 20 * log₁₀(v₁⁄v₂)

Transmitted

Bits: 0

0

1

0

1

0

1

Differential

Voltage

v₁

v₇

v₂

v₃

0V

Time [UI]

v₅

v₆

v₄

v₈

图 8. Conceptual FIR Waveform With Post-Cursor Only

20

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Transmitted

Bits: 0

0

1

0

1

0

1

Differential

Voltage

v₃

v₇

v₁

v₂

0V

Time [UI]

v₄

v₅

v₆

v₈

图 9. Conceptual FIR Waveform With Pre-Cursor Only

Transmitted

Bits: 0

0

1

0

1

0

1

Differential

Voltage

v₇

v₁

v₃

v₂

0V

Time [UI]

v₅

v₆

v₄

v₈

图 10. Conceptual FIR Waveform With Both Pre- and Post-Cursor

8.3.9.1 Setting the Output V_OD, Pre-Cursor, and Post-Cursor Equalization

The output differential voltage (V_OD), pre-cursor, and post-cursor equalization of the driver is controlled by

manipulating the FIR tap settings. The main cursor tap is the primary knob for amplitude adjustment. The pre-

and post-cursor FIR tap settings can then be adjusted to provide equalization. To maintain a constant peak-to-

peak VOD, the user should adjust the main cursor tap value relative to the pre- and post-cursor tap changes so

as to maintain a constant absolute sum of the FIR tap values. The table below shows various settings for V_OD

settings ranging from 205 mVpp to 1225 mVpp (typical). Note that the output peak-to-peak amplitude is a

function of the sum of the absolute values of the taps, whereas the low-frequency amplitude is purely a function

of the main-cursor value.

21

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

RPST(dB)

表 2. Typical VOD and FIR Values

FIR SETTINGS

Peak-to Peak

RPRE(dB)

PRE-CURSOR:

REG_0x3E[6:0]

MAIN-CURSOR:

REG_0x3D[6:0]

POST-CURSOR:

VOD(V)

REG_0x3F[6:0]

0

-1

-2

-3

-4

0

0.205

0.260

0.305

0.355

0.395

0.440

0.490

0.525

0.565

0.610

0.650

0.685

0.720

0.760

0.790

0.825

0.860

0.890

0.925

0.960

0.985

1.010

1.040

1.075

1.095

1.125

1.150

1.165

1.190

1.205

1.220

1.225

0.960

1.165

NA

1.0

1.6

2.4

3.3

NA

2.1

2.5

3.1

3.8

4.7

5.8

7.2

9.0

11.6

NA

1.1

+1

+2

0

+3

0

+4

0

+5

0

+6

0

+7

0

+8

0

+9

0

+10

+11

+12

+13

+14

+15

+16

+17

+18

+19

+20

+21

+22

+23

+24

+25

+26

+27

+28

+29

+30

+31

+18

+17

+16

+15

+14

+13

+12

+11

+10

18

0

-1

-2

-3

-4

-5

-6

-7

-8

-9

0

17

0

16

0

15

0

26

-1

22

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

表 2. Typical VOD and FIR Values (接下页)

FIR SETTINGS

Peak-to Peak

VOD(V)

RPRE(dB)

RPST(dB)

PRE-CURSOR:

MAIN-CURSOR:

REG_0x3D[6:0]

POST-CURSOR:

REG_0x3F[6:0]

REG_0x3E[6:0]

0

25

24

23

22

21

20

19

18

17

16

15

26

25

24

23

22

21

20

-2

-3

-4

-5

-6

-7

-8

-9

-10

-11

-12

0

1.165

NA

0.7

1.2

1.5

2.0

2.6

3.2

4.0

1.3

1.8

2.2

2.7

3.3

3.9

4.7

5.7

6.9

8.4

10.1

NA

0

-1

-2

-3

-4

-5

-6

-7

0

The recommended pre-cursor and post-cursor settings for a given channel will depend on the channel

characteristics (mainly insertion loss) as well as the equalization capabilities of the downstream receiver. The

DS250DF410 receiver, with its highly-capable CTLE and DFE, does not require a significant amount of pre- or

post-cursor. The figures below give general recommendations for pre- and post-cursor for different channel loss

conditions. The insertion loss (IL) in these plots refers to the total loss between the link partner transmitter and

the DS250DF410 receiver.

23

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

图 11. Guideline for Link Partner FIR Settings When IL ≤ 15 dB

图 12. Guideline for Link Partner FIR Settings When IL ≤ 25 dB

24

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

图 13. Guideline for Link Partner FIR Settings When IL ≤ 35 dB

8.3.9.2 Output Driver Polarity Inversion

In some applications, it may be necessary to invert the polarity of the data transmitted from the retimer. To invert

the polarity of the data, read back the FIR polarity settings for the pre, main and post cursor taps and then invert

these bits.

8.3.10 Debug Features

8.3.10.1 Pattern Generator

Each channel in the DS250DF410 can be configured to generate a 16-bit user-defined data pattern or a pseudo

random bit sequence (PRBS). The user defined pattern can also be set to automatically invert every other 16-bit

symbol for DC balancing purposes. The DS250DF410 pattern generator supports the following PRBS

sequences:

•

PRBS – 2⁷- 1

PRBS – 2⁹- 1

PRBS – 2¹¹- 1

PRBS – 2¹⁵- 1

PRBS – 2²³- 1

PRBS – 2³¹- 1

PRBS – 2⁵⁸- 1

PRBS – 2⁶³- 1

25

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

8.3.10.2 Pattern Checker

The pattern checker can be manually set to look for specific PRBS sequences and polarities or it can be set to

automatically detect the incoming pattern and polarity. The PRBS checker supports the same set of PRBS

patterns as the PRBS generator.

The pattern checker consists of an 11-bit error counter. The pattern checker uses 32- bit words, but every bit in

the word is checked for error, so the error count represents the count of single bit errors.

To read out the bit and error counters, the pattern checker must first be frozen. Continuous operation with

simultaneous read out of the bit and error counters is not supported in this implementation. Once the bit and

error counter is read, they can be un-frozen to continue counting.

8.3.10.3 Eye Opening Monitor

The DS250DF410’s Eye Opening Monitor (EOM) measures the internal data eye at the input of the decision

slicer and can be used for 2 functions:

1. Horizontal Eye Opening (HEO) and Vertical Eye Opening (VEO) measurement

2. Full Eye Diagram Capture

The HEO measurement is made at the 0V crossing and is read in channel register 0x27. The VEO measurement

is made at the 0.5 UI mark and is read in channel register 0x28. The HEO and VEO registers can be read from

channel registers 0x27 and 0x28 at any time while the CDR is locked. The following equations are used to

convert the contents of channel registers 0x27 and 0x28 into their appropriate units:

•

HEO [UI] = Reg_0x27 ÷ 32

VEO [mV] = Reg_0x28 x 3.125

A full eye diagram capture can be performed when the CDR is locked. The eye diagram is constructed within a

64 x 64 array, where each cell in the matrix consists of an 16-bit word representing the total number of hits

recorded at that particular phase and voltage offset. Users can manually adjust the vertical scaling of the EOM or

allow the state machine to control the scaling which is the default option. The horizontal scaling controlled by the

state machine and is always directly proportional to the data rate.

When a full eye diagram plot is captured, the retimer will shift out four 16-bit words of residual data that should

be discarded followed by 4096 16-bit words that make up the 64 × 64 eye plot. The first actual word of the eye

plot from the retimer is for (X, Y) position (0,0), which is the earliest position in time and the most negative

position in voltage. Each time the eye plot data is read out the voltage position is incremented. Once the voltage

position has incremented to position 63 (the most positive voltage), the next read will cause the voltage position

to reset to 0 (the most negative voltage) and the phase position to increment. This process will continue until the

entire 64 × 64 matrix is read out. 图 14 below shows the EOM read out sequence overlaid on top of a simple eye

opening plot. In this plot any hits are shown in green. This type of plot is helpful for quickly visualizing the HEO

and VEO. Users can apply different algorithms to the output data to plot density or color gradients to the output

data.

26

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

63

127

4095

63

0

64

4032

63

Phase Position

图 14. EOM Full Eye Capture Readout

To manually control the EOM vertical range, remove scaling control from the state machine then select the

desired range:

Channel Reg 0x2C[6] → 0 (see 表 3).

表 3. Eye Opening Monitor Vertical Range Settings

CH REG 0x11[7:6] VALUE

EOM VERTICAL RANGE [mV]

2’b00

2'b01

2'b10

2'b11

±100

±200

±300

±400

The EOM operates as an under-sampled circuit. This allows the EOM to be useful in identifying over

equalization, ringing and other gross signal conditioning issues. However, the EOM cannot be correlated to a bit

error rate.

The EOM can be accessed in two ways to read out the entire eye plot:

•

Multi-byte reads can be used such that data is repeatedly latched out from channel register 0x25.

With single byte reads, the MSB are located in register 0x25 and the LSB are located in register 0x26. In this

mode, the device must be addressed each time a new byte is read.

To perform a full eye capture with the EOM, follow these steps below within the desired channel register set:

表 4. Eye Opening Monitor Full Eye Capture Instructions

STEP

REGISTER [bits]

0x67[5]

Operation

Write

VALUE

DESCRIPTION

Disable lock EOM lock monitoring

Set the desired EOM vertical range

1

0

0x2C[6]

Write

2

0x11[7:6]

Write

2'b--

27

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

表 4. Eye Opening Monitor Full Eye Capture Instructions (接下页)

STEP

REGISTER [bits]

0x11[5]

Operation

Write

VALUE

DESCRIPTION

3

4

0

1

Power on the EOM

Enable fast EOM

0x24[7]

Write

Begin read out of the 64 x 64 array, discard first 4 words

Ch reg 0x24[0] is self-clearing.

0x24[0]

0x25

0x26

5

6

Read

1

0x25 is the MSB of the 16-bit word

0x26 is the LSB of the 16-bit word

0x25

0x26

Continue reading information until the 64 x 64 array is

complete.

Read

0x67[5]

0x2C[6]

0x11[5]

0x24[7]

Write

1

0

Return the EOM to its original state. Undo steps 1-4

7

8.3.11 Interrupt Signals

The DS250DF410 can be configured to report different events as interrupt signals. These interrupt signals do not

impact the operation of the device, but merely report that the selected event has occurred. The interrupt bits in

the register sets are all sticky bits. This means that when an event triggers an interrupt the status bit for that

interrupt is set to logic HIGH. This interrupt status bit will remain at logic HIGH until the bit has been read. Once

the bit has been read it will be automatically cleared, which allows for new interrupts to be detected. The

DS250DF410 will report the occurrence of an interrupt through the INT_N pin. The INT_N pin is an open drain

output that will pull the line low when an interrupt signal is triggered.

Note that all available interrupts are disabled by default. Users must activate the various interrupts before they

can be used.

The interrupts available in the DS250DF410 are:

•

CDR loss of lock

CDR locked

Signal detect loss

Signal detected

PRBS pattern checker bit error detected

HEO/VEO threshold violation

When an interrupt occurs, share register 0x08 reports which channel generated the interrupt request. Users can

then select the channel(s) that generated the interrupt request and service the interrupt by reading the

appropriate interrupt status bits in the corresponding channel registers. For more information on reading interrupt

status, refer to the DS250DF410 Programming Guide.

8.3.12 JTAG Boundary Scan

The DS250DF410 supports JTAG and AC-JTAG boundary scan to facilitate in-system manufacturing tests. JTAG

mode is enabled with the EN_SMB as shown in 表 5. Refer to the DS250DF410 Product Folder for the boundary

scan description language (BSDL) file for the DS250DF410.

表 5. JTAG Mode pins

INTERNAL PULL-UP /

PULL-DOWN

JTAG MODE FUNCTIONALITY

(EN_SMB = 1 kΩ to GND)

NAME

NO.

F1

CAL_CLK_IN/JTAG_TDI

CAL_CLK_OUT/JTAG_TDO

ADDR1/JTAG_TRS

Weak pull-up

None

JTAG Test Data In (TDI)

JTAG Test Data Out (TDO)

JTAG Test Rset (TRS)

F11

D1

Weak pull-up

None

TEST0/JTAG_TCK

D10

D2

JTAG Test Clock (TCK)

TES1/JTAG_TMS

Weak pull-up

JTAG Test Mode Select (TMS)

28

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

8.4 Device Functional Modes

8.4.1 Supported Data Rates

The DS250DF410 supports a wide range of input data rates, including divide-by-2 and divide-by-4 sub-rates. The

supported data rates are listed in 表 6. Refer to the DS250DF410 Programming Guide for information on

configuring the DS250DF410 for different data rates.

表 6. Supported Data Rates

DATA RATE RANGE

DIVIDER

CDR MODE

COMMENT

MIN

MAX

≥ 20.6 Gbps

≥ 10.3 Gbps

> 6.45 Gbps

≥ 5.15 Gbps

≥ 1.25 Gbps

≤ 25.8 Gbps

≤ 12.9 Gbps

< 10.3 Gbps

≤ 6.45 Gbps

< 5.15 Gbps

1

2

Enabled

Disabled

Enabled

Disabled

N/A

4

Output jitter will be higher with CDR disabled.

N/A

8.4.2 SMBus Master Mode

SMBus master mode allows the DS250DF410 to program itself by reading directly from an external EEPROM.

When using the SMBus master mode, the DS250DF410 will read directly from specific location in the external

EEPROM. When designing a system for using the external EEPROM, the user needs to follow these specific

guidelines:

•

Maximum EEPROM size is 2048 Bytes

Minimum EEPROM size for a single DS250DF410 with individual channel configuration is 305 Bytes (3 base

header bytes + 12 address map bytes + 4 × 72 channel register bytes + 2 share register bytes; bytes are

defined to be 8-bits)

•

Set ENSMB = Float, for SMBus master mode

The external EEPROM device address byte must be 0xA0

The external EEPROM device must support 400kHz operation at 2.5-V or 3.3-V supply

Set the SMBus address of the DS250DF410 by configuring the ADDR0 and ADDR1 pins

When loading multiple DS250DF410 devices from the same EEPROM, use these guidelines to configure the

devices:

•

Configure the SMBus addresses for each DS250DF410 to be sequential. The first device in the sequence

must have an address of 0x30

Daisy chain READ_EN_N and ALL_DONE_N from one device to the next device in the sequence so that they

do not compete for the EEPROM at the same time.

If all of the DS250DF410 devices share the same EEPROM channel and share register settings, configure

the common channel bit in the base header to 1. With common channel configuration enabled, each

DS250DF410 device will configure all 4 channels with the same settings.

When loading a single DS250DF410 from an EEPROM, use these guidelines to configure the device:

•

Set the common channel bit to 0 to allow for individual channel configuration, or set the common channel bit

to 1 to load the same configuration settings to all channels.

•

When configuring individual channels, a 512, 1024 or 2048 Byte EEPROM must be used.

If there are more than three DS250DF410 devices on a PCB that require individual channel configuration,

then each device must have its own EEPROM.

29

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

8.4.3 Device SMBus Address

The DS250DF410’s SMBus slave address is strapped at power up using the ADDR[1:0] pins. The pin state is

read on power up, after the internal power-on reset signal is de-asserted. The ADDR[1:0] pins are four-level

LVCMOS IOs, which provides for 16 unique SMBus addresses. The four levels are achieved by pin strap options

as follows:

•

0: 1 kΩ to GND

R: 10 kΩ to GND (20 kΩ also acceptable)

F: Float

1: 1 kΩ to VDD

表 7. SMBus Address Map

REQUIRED ADDRESS PIN STRAP VALUE

8-BIT WRITE ADDRESS [HEX]

ADDR1

ADDR0

0x30

0x32

0x34

0x36

0x38

0x3A

0x3C

0x3E

0x40

0x42

0x44

0x46

0x48

0x4A

0x4C

0x4E

0

R

F

1

0

R

F

1

0

R

F

1

0

R

F

1

0

1

R

F

1

8.5 Programming

8.5.1 Bit Fields in the Register Set

Many of the registers in the DS250DF410 are divided into bit fields. This allows a single register to serve multiple

purposes which may be unrelated. Often, configuring the DS250DF410 requires writing a bit field that makes up

only part of a register value while leaving the remainder of the register value unchanged. The procedure for

accomplishing this task is to read in the current value of the register to be written, modify only the desired bits in

this value, and write the modified value back to the register. Of course, if the entire register is to be changed,

rather than just a bit field within the register, it is not necessary to read in the current value of the register first. In

all register configuration procedures described in the following sections, this procedure should be kept in mind. In

some cases, the entire register is to be modified. When only a part of the register is to be changed, however, the

procedure described above should be used.

Most register bits can be read or written to. However, some register bits are constrained to specific interface

instructions.

Register bits can have the following interface constraints:

•

R - Read only

RW - Read/Write

RWSC - Read/Write, self-clearing

30

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Programming (接下页)

8.5.2 Writing to and Reading from the Global/Shared/Channel Registers

The DS250DF410 has 3 types of registers:

1) Global Registers – These registers can be accessed at any time and are used to select individual channel

registers, the shared registers or to read back the TI ID and version information.

2) Shared Registers – These registers are used for device-level configuration, status read back or control.

3) Channel Registers – These registers are used to control and configure specific features for each individual

channel. All channels have the same channel register set and can be configured independent of each other.

The global registers can be accessed at any time, regardless of whether the shared or channel register set is

selected. The DS250DF410 global registers are located on addresses 0xEF-0xFF. The function of the global

registers falls into the following categories:

•

Channel selection and share enabling – Registers 0xFC and 0xFF

Device and version information – Registers 0xEF-0xF3

Reserved/unused registers – all other addresses

Register 0xFC is used to select the channel registers to be written to. To select a channel, write a 1 to its

corresponding bit in register 0xFC. Note that more than one channel may be written to by setting multiple bits in

register 0xFC. However, when performing an SMBus read transaction only one channel can be selected at a

time. If multiple channels are selected in register 0xFC when attempting to perform an SMBus read, the device

will return 0xFF.

Register 0xFF bit 1 can be used to perform broadcast register writes to all channels. A single channel read-

modify broadcast write type commands can be accomplished by setting register 0xFF to 0x03 and selecting a

single channel in register 0xFC. This type of configuration allows for the reading of a single channel's register

information and then writing to all channels with the modified value. Register 0xFF bit 0 is used to select the

shared register page or the channel register page for the channels selected in register 0xFC.

TI repeaters/retimers have a vendor ID register (0xFE) which will always read back 0x03. In addition, there are

three device ID registers (0xF0, 0xF1, and 0xF3). These are useful to verify that there is a good SMBus

connection between the SMBus master and the DS250DF410.

31

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

8.6 Register Maps

Table 8. Global Registers

DEFAULT

ADDRESS

(HEX)

BITS

VALUE

(HEX)

MODE

EEPROM

FIELD NAME

DESCRIPTION

EF

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

0

1

0

1

0

1

0

1

0

1

0

R

N

SPARE

R

CHAN_CONFIG_ID[3]

CHAN_CONFIG_ID[2]

CHAN_CONFIG_ID[1]

CHAN_CONFIG_ID[0]

VERSION[7]

TI device ID (Quad count).

DS250DF810: 0x0C

DS250DF410: 0x0E

R

F0

F1

F3

FB

FC

R

Version ID

R

VERSION[6]

R

VERSION[5]

R

VERSION[4]

R

VERSION[3]

R

VERSION[2]

R

VERSION[1]

R

VERSION[0]

R

DEVICE_ID[7]

DEVICE_ID[6]

DEVICE_ID[5]

DEVICE_ID[4]

DEVICE_ID[3]

DEVICE_ID[2]

DEVICE_ID[1]

DEVICE_ID[0]

CHAN_VERSION[3]

CHAN_VERSION[2]

CHAN_VERSION[1]

CHAN_VERSION[0]

SHARE_VERSION[3]

SHARE_VERSION[2]

SHARE_VERSION[1]

SHARE_VERSION[0]

RESERVED

Full device ID

R

Digital Channel Version

Digital Share Version

R

RW

RESERVED

EN_CH7

Select channel 7 (DS250DF810 only)

Select channel 6 (DS250DF810 only)

Select channel 5 (DS250DF810 only)

Select channel 4 (DS250DF810 only)

EN_CH6

EN_CH5

EN_CH4

32

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Register Maps (continued)

Table 8. Global Registers (continued)

DEFAULT

ADDRESS

(HEX)

BITS

VALUE

(HEX)

MODE

EEPROM

FIELD NAME

DESCRIPTION

Select channel 3

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

0

1

0

RW

R

N

EN_CH3

EN_CH2

EN_CH1

EN_CH0

Select channel 2

Select channel 1

Select channel 0

RESERVED

TI vendor ID

FD

RESERVED

FE

VENDOR_ID[7]

VENDOR_ID[6]

VENDOR_ID[5]

VENDOR_ID[4]

VENDOR_ID[3]

VENDOR_ID[2]

VENDOR_ID[1]

VENDOR_ID[0]

RESERVED

R

FF

RW

RESERVED

EN_SHARE_Q1

Select shared registers for quad 1

(DS250DF810 only)

5

1

RW

N

4

3

2

0

RW

N

EN_SHARE_Q0

RESERVED

Select shared registers for quad 0

RESERVED

WRITE_ALL_CH

Allows user to write to all channels as

if they are the same, but only allows

read back from the channel specified

in 0xFC.

1

0

RW

N

Note: EN_CH_SMB must be = 1 or

else this function is invalid.

EN_CH_SMB

1: Enables SMBUS access to the

channels specified in Reg_0xFC

0: The shared registers are selected

33

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 9. Shared Registers

DEFAULT

ADDRESS

(HEX)

FIELD

BITS

VALUE

(HEX)

MODE

EEPROM

NAME

DESCRIPTION

SMBus Address

00

7

6

1

R

N

SMBUS_ADDR[3]

SMBUS_ADDR[2]

Strapped 7-bit addres is 0x18 +

SMBus_Addr[3:0]

1

5

4

0

1

0

1

0

1

0

R

N

SMBUS_ADDR[1]

SMBUS_ADDR[0]

RESERVED

RST_I2C_REGS

3:0

7

R

RESERVED

01

R

6

R

5

R

4

R

3

R

2

R

1

R

0

R

02

03

04

7:0

7

RW

1: Reset shared registers. This bit is

self-clearing.

6

0

RWSC

N

0: Normal operation

RST_I2C_MAS

1: Reset for SMBus/I2C Master. This

bit is self-clearing.

5

4

0

RWSC

RW

N

0: Normal operation

FRC_EEPRM_RD

1: Force EEPROM Configuration

0: Normal operation

3

2

1

0

1

0

1

RW

Y

N

RESERVED

05

DISAB_EEPRM_CFG

1: Disable Master Mode EEPROM

configuration (if not started; this bit is

not effective if EEPROM configuration

is already started)

7

0

RW

N

0: Normal operation

6:5

4

0

1

RW

R

N

RESERVED

EEPROM_READ_DONE

1: SMBus Master mode EEPROM

read complete

0: SMBus Master mode EEPROM

read not started or not complete

TEST0_AS_CAL_CLK_IN

CAL_CLK_INV_DIS

1: Use TEST0 as the input for the

25MHz CAL_CLK instead of

CAL_CLK_IN. This must be configured

for quad0 only.

0: Normal operation. Use

CAL_CLK_IN as the input for the

25MHz CAL_CLK.

3

0

RW

N

1: Disable the inversion of

CAL_CLK_OUT

2

0

RW

Y

0: Normal operation. CAL_CLK_OUT

is inverted with respect to

CAL_CLK_IN.

1

0

1

RW

Y

RESERVED

34

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 9. Shared Registers (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

FIELD

NAME

BITS

MODE

EEPROM

DESCRIPTION

06

08

7:0

7

0

RW

R

N

RESERVED

INT_Q0C3

RESERVED

6

R

5

R

4

R

Interrupt from channel 3. For

DS250DF810, this applies to the quad

selected by Reg_0xFF[5:4].

3

2

1

0

R

N

INT_Q0C2

INT_Q0C1

INT_Q0C0

RESERVED

Interrupt from channel 2. For

DS250DF810, this applies to the quad

selected by Reg_0xFF[5:4].

Interrupt from channel 1. For

DS250DF810, this applies to the quad

selected by Reg_0xFF[5:4].

Interrupt from channel 0. For

DS250DF810, this applies to the quad

selected by Reg_0xFF[5:4].

0

R

N

Y

0A

0B

7:1

RESERVED

DIS_REFCLK_OUT

1: Disable CAL_CLK_OUT (high-Z)

0

RW

Y

0: Normal operation. Enable

CAL_CLK_OUT

7

6

0

RW

R

N

RESERVED

REFCLK_DET

1: 25MHz clock detected on

CAL_CLK_IN

0: No clock detected on CAL_CLK_IN

5

4

0

RW

N

RESERVED

MR_REFCLK_DET_DIS

0: CAL_CLK_IN detection and status

reporting enabled (default)

3

0

RW

N

1: CAL_CLK_IN detection disabled

2

1

0

1

0

RW

R

N

Y

N

RESERVED

EECFG_CMPLT

EECFG_FAIL

RESERVED

0

0C

0D

0E

7:0

7:2

1:0

7:0

7

RW

R

0F

10

RW

R

6

5

4

3

2

1

0

11

7

11: Not valid

10: EEPROM load completed

successfully

01: EEPROM load failed after 64

attempts

6

0

R

N

00: EEPROM load in progress

35

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 9. Shared Registers (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

FIELD

NAME

BITS

MODE

EEPROM

DESCRIPTION

5

4

3

2

1

0

R

N

EECFG_ATMPT[5]

EECFG_ATMPT[4]

EECFG_ATMPT[3]

EECFG_ATMPT[2]

EECFG_ATMPT[1]

EECFG_ATMPT[0]

REG_I2C_FAST

Number of attempts made to load

EEPROM image

12

1: EEPROM load uses Fast I2C Mode

(400 kHz)

7

1

RW

N

0: EEPROM load uses Standard I2C

Mode (100 kHz)

6

5

4

3

2

1

0

1

0

1

RW

N

RESERVED

36

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

00

7

6

5

4

3

0

RW

N

RESERVED

RST_CORE

1: Reset the 10M core clock domain.

This is the main clock domain for all

the state machines

0: Normal operation

2

1

0

RW

N

RST_REGS

RST_VCO

1: Reset channel registers to power-up

defaults.

0: Normal operation

1: Resets the CDR S2P clock domain,

includes PPM counter, EOM counter.

0: Normal operation

RST_REFCLK

1: Resets the 25MHz reference clock

domain, includes PPM counter. Does

not work if 25MHz clock is not present.

0: Normal operation

01

7

6

0

R

N

SIGDET

Raw Signal Detect observation

POL_INV_DET

Indicates PRBS checker detected

polarity inversion in the locked data

sequence.

5

4

0

R

N

CDR_LOCK_LOSS_INT

PRBS_SEQ_DET[3]

1: Indicates loss of CDR lock after

having acquired it. Bit clears on read.

Feature must be enabled with

Reg_0x31[1]

Indicates the pattern detected on the

input serial stream

0xxx: No detect

3

2

1

0

R

N

PRBS_SEQ_DET[2]

PRBS_SEQ_DET[1]

PRBS_SEQ_DET[0]

1000: 7 bits PRBS sequence

1001: 9 bits PRBS sequence

1010: 11 bits PRBS sequence

1011: 15 bits PRBS sequence

1100: 23 bits PRBS sequence

1101: 31 bits PRBS sequence

1110: 58 bits PRBS sequence

1111: 63 bits PRBS sequence

0

7

0

R

N

SIG_DET_LOSS_INT

CDR_STATUS[7]

Loss of signal indicator, set once

signal is acquired and then lost. Clears

on read. Feature must be enabled with

reg_31[0]

02

This register is used to read the status

of internal signal.

Select what is observable on this bus

using Reg_0x0C[7:4]

6

5

4

3

2

1

0

R

N

CDR_STATUS[6]

CDR_STATUS[5]

CDR_STATUS[4]

CDR_STATUS[3]

CDR_STATUS[2]

CDR_STATUS[1]

CDR_STATUS[0]

37

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

03

04

05

06

07

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

RW

Y

N

EQ_BST0[1]

This register can be used to force an

EQ boost setting if used in conjunction

with channel Reg_0x2D[3].

EQ_BST0[0]

EQ_BST1[1]

EQ_BST1[0]

EQ_BST2[1]

EQ_BST2[0]

EQ_BST3[1]

EQ_BST3[0]

RESERVED

38

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

08

7

6

5

4

3

2

1

0

7

0

1

0

1

0

RW

Y

RESERVED

REG_VCO_CAP_OV

09

Enable bit to override cap_cnt with

value in Reg_0x0B[4:0]

6

0

RW

Y

REG_SET_CP_LVL_LPF_OV

Enable bit to override lpf_dac_val with

value in Reg_0x1F[4:0]

5

4

0

RW

Y

REG_BYPASS_PFD_OV

0: Normal operation.

REG_EN_FD_PD_VCO_PDIQ_ Enable bit to override en_fd, pd_pd,

OV

pd_vco, pd_pdiq with Reg_0x1E[0],

Reg_0x1E[2], Reg_0x1C[0],

Reg_0x1C[1]

3

2

0

RW

Y

REG_EN_PD_CP_OV

REG_DIVSEL_OV

Enable bit to override pd_fd_cp and

pd_pd_cp with value in Reg_0x1B[1:0]

Enable bit to override divsel with value

in Reg_0x18[6:4]

1

0

7

6

0

RW

Y

RESERVED

0A

RESERVED

REG_EN_IDAC_PD_CP_OV_

Enable bit to override phase detector

AND_REG_EN_IDAC_FD_CP_ charge pump settings with

OV

Reg_0x1C[7:5]

Enable bit to override frequency

detector charge pump settings with

Reg_0x1C[4:2]

5

0

RW

Y

REG_DAC_LPF_HIGH_PHASE_ Enable bit to loop filter comparator trip

OV_

voltages with Reg_0x16[7:0]

AND_REG_DAC_LPF_LOW_PH

ASE_OV

4

3

0

RW

Y

N

RESERVED

REG_CDR_RESET_OV

Enable CDR Reset override with

Reg_0x0A[2]

2

1

0

RW

N

REG_CDR_RESET_SM

REG_CDR_LOCK_OV

CDR Reset override bit

Enable CDR lock signal override with

Reg_0x0A[0]

0

7

6

5

4

3

2

1

0

1

0

1

RW

N

Y

REG_CDR_LOCK

RESERVED

CDR lock signal override bit

RESERVED

0B

RESERVED

39

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

0C

7

6

5

4

3

2

1

0

7

0

1

RW

N

RESERVED

DES_PD

0D

1: De-serializer (for PRBS checker) is

powered down

0: De-serializer (for PRBS checker) is

enabled

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

1

0

1

0

RW

N

Y

N

RESERVED

0E

0F

10

40

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

11

7

6

0

RW

Y

EOM_SEL_VRANGE[1]

EOM_SEL_VRANGE[0]

Manually set the EOM vertical range,

used with channel Reg_0x2C[6]:

00: ±100 mV

01: ±200 mV

10: ±300 mV

11: ±400 mV

5

1

RW

Y

EOM_PD

1: Normal operation. Eye opening

monitor (EOM) is automatically duty-

cycled.

0: EOM is force-enabled

4

3

0

RW

N

Y

RESERVED

DFE_TAP2_POL

Bit forces DFE tap 2 polarity

1: Negative, boosts by the specified

tap weight

0: Positive, attenuates by the specified

tap weight

2

1

0

7

0

1

RW

Y

DFE_TAP3_POL

DFE_TAP4_POL

DFE_TAP5_POL

DFE_TAP1_POL

Bit forces DFE tap 3 polarity

1: Negative, boosts by the specified

tap weight

0: Positive, attenuates by the specified

tap weight

Bit forces DFE tap 4 polarity

1: Negative, boosts by the specified

tap weight

0: Positive, attenuates by the specified

tap weight

Bit forces DFE tap 5 polarity

1: Negative, boosts by the specified

tap weight

0: Positive, attenuates by the specified

tap weight

12

Bit forces DFE tap 1 polarity

1: Negative, boosts by the specified

tap weight

0: Positive, attenuates by the specified

tap weight

6

5

4

3

2

1

0

1

RW

N

Y

RESERVED

DFE_WT1[4]

DFE_WT1[3]

DFE_WT1[2]

DFE_WT1[1]

DFE_WT1[0]

RESERVED

These bits force DFE tap 1 weight.

Manual DFE operation is required for

this to take effect by setting

Reg_0x15[7]=1.

If Reg_0x15[7]=0, the value defined

here is used as the initial DFE tap 1

weight during adaptation.

41

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

13

7

1

0

1

RW

N

EQ_PD_PEAKDETECT

1: Normal operation. Power down test

mode.

0: Test mode.

6

5

RW

Y

EQ_PD_SD

1: Power down signal detect.

0: Normal operation. Enable signal

detect.

EQ_HI_GAIN

1: Enable high DC gain mode in the

equalizer

0: Enable low DC gain mode in the

equalizer

(Refer to the Programming Guide for

more details)

4

1

RW

Y

EQ_EN_DC_OFF

1: Normal operation.

0: Disable DC offset compensation.

3

2

0

RW

Y

RESERVED

EQ_LIMIT_EN

1: Configures the final stage of the

equalizer to be a limiting stage.

0: Normal operation, final stage of the

equalizer is configured to be a non-

limiting stage.

1

0

7

0

RW

Y

RESERVED

14

EQ_SD_PRESET

1: Forces signal detect HIGH, and

force enables the channel. Should not

be set if bit 6 is set.

0: Normal Operation.

6

0

RW

Y

EQ_SD_RESET

1: Forces signal detect LOW and force

disables the channel. Should not be

set if bit 7 is set.

0: Normal Operation.

5

4

0

RW

Y

EQ_REFA_SEL1

EQ_REFA_SEL0

Controls the signal detect assert

levels.

(Refer to the Programming Guide for

more details)

3

2

0

1

RW

Y

EQ_REFD_SEL1

EQ_REFD_SEL0

Controls the signal detect de-assert

levels.

(Refer to the Programming Guide for

more details)

1

0

7

0

RW

N

Y

RESERVED

15

DFE_FORCE_EN

1: Enables manual DFE tap settings

0: Normal operation

6

5

4

3

0

1

0

RW

N

Y

RESERVED

DRV_PD

RESERVED

1: Powers down the high speed driver

0: Normal operation

2

1

0

RW

Y

RESERVED

42

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

16

17

18

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

0

1

0

1

0

1

0

1

0

1

0

RW

Y

N

Y

RESERVED

PDIQ_SEL_DIV[2]

PDIQ_SEL_DIV[1]

PDIQ_SEL_DIV[0]

These bits will force the divider setting

if 0x09[2] is set.

000: Divide by 1

001: Divide by 2

010: Divide by 4

011: Divide by 8

100: Divide by 16

All other values are reserved.

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

RW

N

Y

N

Y

N

RESERVED

19

1A

43

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

1B

7

6

5

4

3

2

1

0

1

RW

N

Y

RESERVED

CP_EN_CP_PD

1: Normal operation, phase detector

charge pump enabled

0

1

RW

Y

CP_EN_CP_FD

1: Normal operation, frequency

detector charge pump enabled

1C

1D

1E

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

1

0

1

0

1

RW

Y

N

Y

EN_IDAC_PD_CP2

EN_IDAC_PD_CP1

EN_IDAC_PD_CP0

EN_IDAC_FD_CP2

EN_IDAC_FD_CP1

EN_IDAC_FD_CP0

RESERVED

Phase detector charge pump setting.

Override bit required for these bits to

take effect

Frequency detector charge pump

setting. Override bit required for these

bits to take effect

RESERVED

PFD_SEL_DATA_PRELCK[2]

PFD_SEL_DATA_PRELCK[1]

PFD_SEL_DATA_PRELCK[0]

Output mode for when the CDR is not

locked. For these values to take effect,

Reg_0x09[5] must be set to 0, which is

the default.

000: Raw Data

111: Mute (Default)

All other values are reserved. (Refer to

the Programming Guide for more

details)

4

3

0

1

RW

N

Y

SER_EN

DFE_PD

1: Enable serializer (used for PRBS

Generator)

0: Normal operation. Disable serializer.

This bit must be cleared for the DFE to

be functional in any adapt mode.

1: (Default) DFE disabled.

0: DFE enabled

2

1

0

1

RW

Y

PFD_PD_PD

1: Power down PFD phase detector.

0: Normal operation. Enable PFD

phase detector.

EN_PARTIAL_DFE

PFD_EN_FD

1: Enable DFE taps 3-5. DFE_PD

must also be set to 0.

0: (Default) Disable DFE taps 3-5.

1: Normal operation. Enable PFD

frequency detector.

0: Disable PFD frequency detector.

44

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

1F

7

6

5

4

3

0

1

RW

N

Y

RESERVED

MR_LPF_AUTO_ADJST_EN

1: Normal operation. Allow LPF to tune

to optimum value during fast-cap

search routine.

0: Otherwise LPF value is determined

by the Reg_0x9D.

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

0

1

0

RW

Y

N

RESERVED

DFE_WT5[3]

DFE_WT5[2]

DFE_WT5[1]

DFE_WT5[0]

DFE_WT4[3]

DFE_WT4[2]

DFE_WT4[1]

DFE_WT4[0]

DFE_WT3[3]

DFE_WT3[2]

DFE_WT3[1]

DFE_WT3[0]

DFE_WT2[3]

DFE_WT2[2]

DFE_WT2[1]

DFE_WT2[0]

EOM_OV

RESERVED

20

21

22

Bits force DFE tap 5 weight, manual

DFE operation required to take effect

by setting 0x15[7]=1.

Bits force DFE tap 4 weight, manual

DFE operation required to take effect

by setting 0x15[7]=1.

Bits force DFE tap 3 weight, manual

DFE operation required to take effect

by setting 0x15[7]=1.

Bits force DFE tap 2 weight, manual

DFE operation required to take effect

by setting 0x15[7]=1.

1: Override enable for EOM manual

control

0: Normal operation

6

0

RW

N

EOM_SEL_RATE_OV

1: Override enable for EOM rate

selection

0: Normal operation

5

4

3

2

1

0

RW

N

RESERVED

45

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

23

7

0

RW

N

EOM_GET_HEO_VEO_OV

1: Override enable for manual control

of the HEO/VEO trigger

0: Normal operation

6

1

RW

Y

DFE_OV

1: Normal operation; DFE must be

enabled in Reg_0x1E[3].

5

4

3

2

1

0

7

0

RW

N

RESERVED

FAST_EOM

RESERVED

24

1: Enables fast EOM for full eye

capture. In this mode the phase DAC

and voltage DAC or the EOM are

automatically incremented through a

64 x 64 matrix. Values for each point

are stored in Reg_0x25 and

Reg_0x26.

0: Normal operation.

6

5

0

R

N

DFE_EQ_ERROR_NO_LOCK

DFE/CTLE SM quit due to loss of lock

GET_HEO_VEO_ERROR_NO_ get_heo_veo sees no hits at zero

HITS crossing

4

0

R

N

GET_HEO_VEO_ERROR_NO_ get_heo_veo cannot see a vertical eye

OPENING

opening

3

2

0

RW

N

RESERVED

DFE_ADAPT

RESERVED

RWSC

1: Manually start DFE adaption (self-

clearing).

0: Normal operation.

1

0

R

N

EOM_GET_HEO_VEO

1: Manually triggers HEO/VEO

measurement; feature must be

enabled with Reg_0x23[7]; the

HEO/VEO values are read from

Reg_0x27, Reg_0x28

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

RWSC

R

N

EOM_START

1: Starts EOM counter, self-clearing

MSBs of EOM counter

25

EOM_COUNT15

EOM_COUNT14

EOM_COUNT13

EOM_COUNT12

EOM_COUNT11

EOM_COUNT10

EOM_COUNT9

EOM_COUNT8

EOM_COUNT7

EOM_COUNT6

EOM_COUNT5

EOM_COUNT4

EOM_COUNT3

EOM_COUNT2

EOM_COUNT1

EOM_COUNT0

R

26

R

LSBs of EOM counter

R

46

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

27

28

29

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

0

R

N

HEO7

HEO6

HEO5

HEO4

HEO3

HEO2

HEO1

HEO0

VEO7

VEO6

VEO5

VEO4

VEO3

VEO2

VEO1

VEO0

HEO value, requires CDR to be locked

for valid measurement

R

VEO value, requires CDR to be locked

for valid measurement

R

RW

R

RESERVED

EOM_VRANGE_SETTING[1]

EOM_VRANGE_SETTING[0]

Read the currently set Eye Monitor

Voltage Range:

11 - +/-400mV

R

10 - +/- 300mV

01 - +/- 200mV

00 - +/- 100mV"

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

RW

R

N

Y

RESERVED

VEO[8]

VEO MSB value

HEO MSB value

R

HEO[8]

2A

RW

EOM_TIMER_THR[3]

EOM_TIMER_THR[2]

EOM_TIMER_THR[1]

EOM_TIMER_THR[0]

VEO_MIN_REQ_HITS[3]

VEO_MIN_REQ_HITS[2]

VEO_MIN_REQ_HITS[1]

VEO_MIN_REQ_HITS[0]

The value of EOM_TIMER_THR[7:4]

controls the amount of time the Eye

Monitor samples each point in the eye.

(Refer to the Programming Guide for

more details)

Whenever the Eye Monitor is used to

measure HEO and VEO, the data is

sampled for some number of bits, set

by Reg_0x2A[7:4]. This register sets

the number of hits within that sample

size that is required before the EOM

will indicate a hit has occurred. This

filtering only affects the VEO

measurement.

47

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

2B

7

6

5

4

3

2

1

0

1

0

1

0

RW

N

Y

RESERVED

EOM_MIN_REQ_HITS[3]

EOM_MIN_REQ_HITS[2]

EOM_MIN_REQ_HITS[1]

EOM_MIN_REQ_HITS[0]

Whenever the Eye Monitor is used to

measure HEO and VEO, the data is

sampled for some number of bits, set

by Reg_0x2A[7:4]. This register sets

the number of hits within that sample

size that is required before the EOM

will indicate a hit has occurred. This

filtering only affects the HEO

measurement.

2C

7

6

1

RW

N

Y

RELOAD_DFE_TAPS

VEO_SCALE

Causes DFE taps to load from last

adapted values

1: Normal operation. Scale VEO

based on EOM vrange.

5

4

1

RW

Y

DFE_SM_FOM1

DFE_SM_FOM0

This register defines the Figure of

Merit used when adapting the DFE:

00: not valid

01: SM uses only HEO

10: SM uses only VEO

11: SM uses both HEO and VEO

Additionally, if Reg_0x6E[6] is set to

'1', the Alternate FOM is used. This bit

takes precedence over DFE_SM_FOM

3

2

1

0

7

6

5

4

3

0

1

0

1

0

RW

Y

DFE_ADAPT_COUNTER[3]

DFE_ADAPT_COUNTER[2]

DFE_ADAPT_COUNTER[1]

DFE_ADAPT_COUNTER[0]

RESERVED

DFE look-beyond count.

2D

RESERVED

REG_EQ_BST_OV

1: Allow override control of the EQ

setting by writing to Reg_0x03

0: Normal operation.

2

1

0

7

6

5

0

RW

R

Y

N

RESERVED

2E

RESERVED

EQ_BST3_BIT2_TO_EQ

Read-back of eq_BST3[2] driving the

EQ

4

3

2

0

RW

N

RESERVED

PRBS_PATTERN_SEL[2]

MSB for the PRBS_PATTERN_SEL

field. Lower bits are found on

Reg_0x30[1:0]. Refer to the Reg_0x30

description on this table.

1

0

RW

N

RESERVED

48

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

2F

7

6

5

4

3

0

1

0

1

0

RW

Y

RESERVED

RATE[2]

RATE[1]

RATE[0]

INDEX_OV

Configure PPM register and divider for

a standard data rate.

(Refer to the Programming Guide for

more details)

If this bit is 1, then Reg_0x39 is to be

used as 4-bit index to the [15:0] array

of EQ settings. The EQ setting at that

index is loaded to the EQ boost

registers going to the analog and is

used as the starting point for adaption.

2

1

RW

Y

EN_PPM_CHECK

1: (Default) Enable the PPM to be

used as a qualifier when performing

Lock Detect

0: Remove the PPM check as a lock

qualifier.

1

0

RW

Y

N

RESERVED

RWSC

CTLE_ADAPT

1: Re-starts CTLE adaptation, self-

clearing

30

7

6

5

0

RW

N

Y

N

FREEZE_PPM_CNT

EQ_SEARCH_OV_EN

EN_PATT_INV

1: Freeze the PPM counter to allow

safe read asynchronously

1: Enables the EQ 'search" bit to be

forced by Reg_0x13[2]

1: Enable automatic pattern inversion

of successive 16 bit words when using

the "Fixed Pattern" generator option.

4

3

0

RW

N

RELOAD_PRBS_CHKR

PRBS_EN_DIG_CLK

1: Force reload of seed into PRBS

checker LFSR without holding the

checker in reset.

This bit enables the clock to operate

the PRBS generator and/or the PRBS

checher. Toggling this bit is the

primary method to reset the PRBS

pattern generator and PRBS checker.

2

0

RW

N

PRBS_PROGPATT_EN

Enable a fixed data pattern output.

Requires that serializer is enabled with

Reg_0x1E[4]. PRBS generator and

checker should be disabled,

Reg_0x30[3]. The fixed data pattern is

set by Reg_0x7C and Reg_0x97.

Enable inversion of the pattern every

16 bits with Reg_0x30[5].

1

0

RW

N

PRBS_PATTERN_SEL[1]

PRBS_PATTERN_SEL[0]

Selects the pattern output when using

the PRBS generator. Requires the

pattern generator to be configured

properly. The MSB for the

PRBS_PATTERN_SEL field is in

Reg_0x2E[2].

Use Reg_0x30[3] to enable the PRBS

generator.

000: 2^7-1 bits PRBS sequence

001: 2^9-1 bits PRBS sequence

010: 2^11-1 bits PRBS sequence

011: 2^15-1 bits PRBS sequence

100: 2^23-1 bits PRBS sequence

101: 2^31-1 bits PRBS sequence

110: 2^58-1 bits PRBS sequence

111: 2^63-1 bits PRBS sequence

49

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

PRBS_INT_EN

DESCRIPTION

31

7

0

RW

N

1: Enables interrupt for detection of

PRBS errors. The PRBS checker must

be properly configured for this feature

to work.

6

5

0

1

RW

Y

ADAPT_MODE[1]

ADAPT_MODE[0]

00: no adaption

01: adapt CTLE only

10: adapt CTLE until optimal, then

DFE, then CTLE again

11: adapt CTLE until lock, then DFE,

then EQ until optimal

Note: for ADAPT_MODE=2 or 3, the

DFE must be enabled by setting

Reg_0x1E[3]=0 and Reg_0x1E[1]=1.

(Refer to the Programming Guide for

more details)

4

3

0

RW

Y

EQ_SM_FOM[1]

EQ_SM_FOM[0]

CTLE (EQ) adaption state machine

figure of merit.

00: (Default) SM uses both HEO and

VEO

01: SM uses HEO only

10: SM uses VEO only

11: SM uses both HEO and VEO

Additionally, if Reg_0x6E[7]=1, the

Alternate FOM is used. Reg_0x6E[7]

takes precedence over EQ_SM_FOM.

2

1

0

RW

N

Y

RESERVED

CDR_LOCK_LOSS_INT_EN

Enable for CDR Lock Loss Interrupt.

Observable in Reg_0x01[5]

0

RW

Y

SIGNAL_DET_LOSS_INT_EN

Enable for Signal Detect Loss

Interrupt. Observable in Reg_0x01[0]

32

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

RW

Y

HEO_INT_THRESH[3]

HEO_INT_THRESH[2]

HEO_INT_THRESH[1]

HEO_INT_THRESH[0]

VEO_INT_THRESH[3]

VEO_INT_THRESH[2]

VEO_INT_THRESH[1]

VEO_INT_THRESH[0]

HEO_THRESH[3]

These bits set the threshold for the

HEO and VEO interrupt. Each

threshold bit represents 8 counts of

HEO or VEO.

33

In adapt mode 3, the register sets the

minimum HEO and VEO required for

CTLE adaption, before starting DFE

adaption. This can be a max of 15.

HEO_THRESH[2]

HEO_THRESH[1]

HEO_THRESH[0]

VEO_THRESH[3]

VEO_THRESH[2]

VEO_THRESH[1]

VEO_THRESH[0]

50

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

PPM_ERR_RDY

DESCRIPTION

34

7

0

R

N

1: Indicates that a PPM error count is

read to be read from channel

Reg_0x3B and Reg_0x3C

6

0

RW

Y

LOW_POWER_MODE_DISABL By default, all blocks (except signal

E

detect) power down after 100 ms after

signal detect goes low. If set high, all

blocks get powered on after the signal

detect initially goes high.

5

4

1

RW

Y

LOCK_COUNTER[1]

LOCK_COUNTER[0]

After achieving lock, the CDR

continues to monitor the lock criteria. If

the lock criteria fail, the lock is

checked for a total of N number of

times before declaring an out of lock

condition, where N is set by this the

value in these registers, with a max

value of +3, for a total of 4. If during

the N lock checks, lock is regained,

then the lock condition is left HI, and

the counter is reset back to zero.

3

2

1

0

7

6

1

0

RW

Y

DFE_MAX_TAP2_5[3]

DFE_MAX_TAP2_5[2]

DFE_MAX_TAP2_5[1]

DFE_MAX_TAP2_5[0]

DATA_LOCK_PPM[1]

DATA_LOCK_PPM[0]

These four bits are used to set the

maximum value by which DFE taps 2-

5 are able to adapt with each

subsequent adaptation. Same used for

both polarities.

35

Modifies the value of the PPM delta

tolerance from channel Reg_0x64:

00 - ppm_delta[7:0] =1 x

ppm_delta[7:0]

01 - ppm_delta[7:0] =1 x

ppm_delta[7:0] + ppm_delta[3:1]

10 - ppm_delta[7:0] =2 x

ppm_delta[7:0]

11 - ppm_delta[7:0] =2 x

ppm_delta[7:0] + ppm_delta[3:1]

5

0

RW

N

GET_PPM_ERROR

Get PPM error from PPM_COUNT -

clears when done. Normally updates

continuously, but can be manually

triggered with read value from

Reg_0x3B and Reg_0x3C

4

3

2

1

0

7

6

0

1

0

RW

Y

N

Y

DFE_MAX_TAP1[4]

DFE_MAX_TAP1[3]

DFE_MAX_TAP1[2]

DFE_MAX_TAP1[1]

DFE_MAX_TAP1[0]

RESERVED

Limits DFE tap 1 maximum magnitude.

36

RESERVED

HEO_VEO_INT_EN

1: Enable HEO/VEO interrupt

capability

5

4

3

2

1

0

1

0

RW

Y

N

Y

N

REF_MODE[1]

REF_MODE[0]

RESERVED

11: Normal Operation. Refererence

mode 3.

RESERVED

51

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 10. Channel Registers, 0 to 39 (continued)

DEFAULT

VALUE

(HEX)

ADDRESS

(HEX)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

37

38

39

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

0

1

R

N

Y

CTLE_STATUS[7]

Feature is reserved for future use

RESERVED

CTLE_STATUS[6]

CTLE_STATUS[5]

CTLE_STATUS[4]

CTLE_STATUS[3]

CTLE_STATUS[2]

CTLE_STATUS[1]

CTLE_STATUS[0]

DFE_STATUS[7]

DFE_STATUS[6]

DFE_STATUS[5]

DFE_STATUS[4]

DFE_STATUS[3]

DFE_STATUS[2]

DFE_STATUS[1]

DFE_STATUS[0]

RESERVED

R

RW

MR_EOM_RATE[1]

MR_EOM_RATE[0]

With eom_ov = 1, these bits control

the Eye Monitor Rate:

11: Use for full rate, fastest

10: Use for 1/2 Rate

All other values are reserved

4

3

2

1

0

RW

Y

RESERVED

START_INDEX[3]

START_INDEX[2]

START_INDEX[1]

START_INDEX[0]

Start index for EQ adaptation

52

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

3A

3B

3C

3D

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

0

RW

R

Y

N

Y

FIXED_EQ_BST0[1]

During adaptation, if the divider

setting is >2, then a fixed EQ setting

from this register will be used.

However, if channel Reg_0x6F[7] is

enabled, then an EQ adaptation will

be performed instead

FIXED_EQ_BST0[0]

FIXED_EQ_BST1[1]

FIXED_EQ_BST1[0]

FIXED_EQ_BST2[1]

FIXED_EQ_BST2[0]

FIXED_EQ_BST3[1]

FIXED_EQ_BST3[0]

PPM_COUNT[15]

PPM_COUNT[14]

PPM_COUNT[13]

PPM_COUNT[12]

PPM_COUNT[11]

PPM_COUNT[10]

PPM_COUNT[9]

PPM_COUNT[8]

PPM_COUNT[7]

PPM_COUNT[6]

PPM_COUNT[5]

PPM_COUNT[4]

PPM_COUNT[3]

PPM_COUNT[2]

PPM_COUNT[1]

PPM_COUNT[0]

EN_FIR_CURSOR

PPM count MSB

R

PPM count LSB

R

RW

1: Enable Pre- and Post-cursor FIR

0: Disable Pre- and Post-cursor FIR

(lower power)

6

0

RW

Y

FIR_C0_SGN

Main-cursor sign bit

0: positive

1: negative

5

4

3

2

1

0

7

6

0

1

0

1

0

1

RW

Y

RESERVED

FIR_C0[4]

RESERVED

Main-cursor magnitude

(Refer to the Programming Guide for

more details)

FIR_C0[3]

FIR_C0[2]

FIR_C0[1]

FIR_C0[0]

3E

FIR_PD_TX

FIR_CN1_SGN

Pre-cursor sign bit

1: negative

0: positive

5

4

3

2

1

0

RW

Y

RESERVED

FIR_CN1[3]

FIR_CN1[2]

FIR_CN1[1]

FIR_CN1[0]

RESERVED

Pre-cursor magnitude

(Refer to the Programming Guide for

more details)

53

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

3F

7

6

0

1

RW

Y

RESERVED

FIR_CP1_SGN

Post-cursor sign bit

1: negative

0: positive

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

RW

Y

RESERVED

FIR_CP1[3]

Post-cursor magnitude

(Refer to the Programming Guide for

more details)

FIR_CP1[2]

FIR_CP1[1]

FIR_CP1[0]

40

41

42

43

EQ_ARRAY_INDEX_0_BST0[1]

EQ_ARRAY_INDEX_0_BST0[0]

EQ_ARRAY_INDEX_0_BST1[1]

EQ_ARRAY_INDEX_0_BST1[0]

EQ_ARRAY_INDEX_0_BST2[1]

EQ_ARRAY_INDEX_0_BST2[0]

EQ_ARRAY_INDEX_0_BST3[1]

EQ_ARRAY_INDEX_0_BST3[0]

EQ_ARRAY_INDEX_1_BST0[1]

EQ_ARRAY_INDEX_1_BST0[0]

EQ_ARRAY_INDEX_1_BST1[1]

EQ_ARRAY_INDEX_1_BST1[0]

EQ_ARRAY_INDEX_1_BST2[1]

EQ_ARRAY_INDEX_1_BST2[0]

EQ_ARRAY_INDEX_1_BST3[1]

EQ_ARRAY_INDEX_1_BST3[0]

EQ_ARRAY_INDEX_2_BST0[1]

EQ_ARRAY_INDEX_2_BST0[0]

EQ_ARRAY_INDEX_2_BST1[1]

EQ_ARRAY_INDEX_2_BST1[0]

EQ_ARRAY_INDEX_2_BST2[1]

EQ_ARRAY_INDEX_2_BST2[0]

EQ_ARRAY_INDEX_2_BST3[1]

EQ_ARRAY_INDEX_2_BST3[0]

EQ_ARRAY_INDEX_3_BST0[1]

EQ_ARRAY_INDEX_3_BST0[0]

EQ_ARRAY_INDEX_3_BST1[1]

EQ_ARRAY_INDEX_3_BST1[0]

EQ_ARRAY_INDEX_3_BST2[1]

EQ_ARRAY_INDEX_3_BST2[0]

EQ_ARRAY_INDEX_3_BST3[1]

EQ_ARRAY_INDEX_3_BST3[0]

54

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

44

45

46

47

48

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

RW

Y

EQ_ARRAY_INDEX_4_BST0[1]

EQ_ARRAY_INDEX_4_BST0[0]

EQ_ARRAY_INDEX_4_BST1[1]

EQ_ARRAY_INDEX_4_BST1[0]

EQ_ARRAY_INDEX_4_BST2[1]

EQ_ARRAY_INDEX_4_BST2[0]

EQ_ARRAY_INDEX_4_BST3[1]

EQ_ARRAY_INDEX_4_BST3[0]

EQ_ARRAY_INDEX_5_BST0[1]

EQ_ARRAY_INDEX_5_BST0[0]

EQ_ARRAY_INDEX_5_BST1[1]

EQ_ARRAY_INDEX_5_BST1[0]

EQ_ARRAY_INDEX_5_BST2[1]

EQ_ARRAY_INDEX_5_BST2[0]

EQ_ARRAY_INDEX_5_BST3[1]

EQ_ARRAY_INDEX_5_BST3[0]

EQ_ARRAY_INDEX_6_BST0[1]

EQ_ARRAY_INDEX_6_BST0[0]

EQ_ARRAY_INDEX_6_BST1[1]

EQ_ARRAY_INDEX_6_BST1[0]

EQ_ARRAY_INDEX_6_BST2[1]

EQ_ARRAY_INDEX_6_BST2[0]

EQ_ARRAY_INDEX_6_BST3[1]

EQ_ARRAY_INDEX_6_BST3[0]

EQ_ARRAY_INDEX_7_BST0[1]

EQ_ARRAY_INDEX_7_BST0[0]

EQ_ARRAY_INDEX_7_BST1[1]

EQ_ARRAY_INDEX_7_BST1[0]

EQ_ARRAY_INDEX_7_BST2[1]

EQ_ARRAY_INDEX_7_BST2[0]

EQ_ARRAY_INDEX_7_BST3[1]

EQ_ARRAY_INDEX_7_BST3[0]

EQ_ARRAY_INDEX_8_BST0[1]

EQ_ARRAY_INDEX_8_BST0[0]

EQ_ARRAY_INDEX_8_BST1[1]

EQ_ARRAY_INDEX_8_BST1[0]

EQ_ARRAY_INDEX_8_BST2[1]

EQ_ARRAY_INDEX_8_BST2[0]

EQ_ARRAY_INDEX_8_BST3[1]

EQ_ARRAY_INDEX_8_BST3[0]

55

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

49

4A

4B

4C

4D

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

RW

Y

EQ_ARRAY_INDEX_9_BST0[1]

EQ_ARRAY_INDEX_9_BST0[0]

EQ_ARRAY_INDEX_9_BST1[1]

EQ_ARRAY_INDEX_9_BST1[0]

EQ_ARRAY_INDEX_9_BST2[1]

EQ_ARRAY_INDEX_9_BST2[0]

EQ_ARRAY_INDEX_9_BST3[1]

EQ_ARRAY_INDEX_9_BST3[0]

EQ_ARRAY_INDEX_10_BST0[1]

EQ_ARRAY_INDEX_10_BST0[0]

EQ_ARRAY_INDEX_10_BST1[1]

EQ_ARRAY_INDEX_10_BST1[0]

EQ_ARRAY_INDEX_10_BST2[1]

EQ_ARRAY_INDEX_10_BST2[0]

EQ_ARRAY_INDEX_10_BST3[1]

EQ_ARRAY_INDEX_10_BST3[0]

EQ_ARRAY_INDEX_11_BST0[1]

EQ_ARRAY_INDEX_11_BST0[0]

EQ_ARRAY_INDEX_11_BST1[1]

EQ_ARRAY_INDEX_11_BST1[0]

EQ_ARRAY_INDEX_11_BST2[1]

EQ_ARRAY_INDEX_11_BST2[0]

EQ_ARRAY_INDEX_11_BST3[1]

EQ_ARRAY_INDEX_11_BST3[0]

EQ_ARRAY_INDEX_12_BST0[1]

EQ_ARRAY_INDEX_12_BST0[0]

EQ_ARRAY_INDEX_12_BST1[1]

EQ_ARRAY_INDEX_12_BST1[0]

EQ_ARRAY_INDEX_12_BST2[1]

EQ_ARRAY_INDEX_12_BST2[0]

EQ_ARRAY_INDEX_12_BST3[1]

EQ_ARRAY_INDEX_12_BST3[0]

EQ_ARRAY_INDEX_13_BST0[1]

EQ_ARRAY_INDEX_13_BST0[0]

EQ_ARRAY_INDEX_13_BST1[1]

EQ_ARRAY_INDEX_13_BST1[0]

EQ_ARRAY_INDEX_13_BST2[1]

EQ_ARRAY_INDEX_13_BST2[0]

EQ_ARRAY_INDEX_13_BST3[1]

EQ_ARRAY_INDEX_13_BST3[0]

56

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

4E

4F

50

51

52

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

RW

Y

N

EQ_ARRAY_INDEX_14_BST0[1]

EQ_ARRAY_INDEX_14_BST0[0]

EQ_ARRAY_INDEX_14_BST1[1]

EQ_ARRAY_INDEX_14_BST1[0]

EQ_ARRAY_INDEX_14_BST2[1]

EQ_ARRAY_INDEX_14_BST2[0]

EQ_ARRAY_INDEX_14_BST3[1]

EQ_ARRAY_INDEX_14_BST3[0]

EQ_ARRAY_INDEX_15_BST0[1]

EQ_ARRAY_INDEX_15_BST0[0]

EQ_ARRAY_INDEX_15_BST1[1]

EQ_ARRAY_INDEX_15_BST1[0]

EQ_ARRAY_INDEX_15_BST2[1]

EQ_ARRAY_INDEX_15_BST2[0]

EQ_ARRAY_INDEX_15_BST3[1]

EQ_ARRAY_INDEX_15_BST3[0]

RESERVED

57

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

53

54

55

56

57

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

RW

N

RESERVED

58

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

58

59

5A

5B

5C

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

RW

N

RESERVED

59

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

5D

5E

5F

60

61

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

RW

N

Y

RESERVED

Group 0 count LSB

RESERVED

GRP0_OV_CNT[7]

GRP0_OV_CNT[6]

GRP0_OV_CNT[5]

GRP0_OV_CNT[4]

GRP0_OV_CNT[3]

GRP0_OV_CNT[2]

GRP0_OV_CNT[1]

GRP0_OV_CNT[0]

CNT_DLTA_OV_0

Override enable for group 0 manual

data rate selection

6

5

4

3

2

1

0

RW

Y

GRP0_OV_CNT[14]

GRP0_OV_CNT[13]

GRP0_OV_CNT[12]

GRP0_OV_CNT[11]

GRP0_OV_CNT[10]

GRP0_OV_CNT[9]

GRP0_OV_CNT[8]

Group 0 count MSB

60

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

Group 1 count LSB

62

7

6

5

4

3

2

1

0

7

0

RW

Y

GRP1_OV_CNT[7]

GRP1_OV_CNT[6]

GRP1_OV_CNT[5]

GRP1_OV_CNT[4]

GRP1_OV_CNT[3]

GRP1_OV_CNT[2]

GRP1_OV_CNT[1]

GRP1_OV_CNT[0]

CNT_DLTA_OV_1

63

Override enable for group 1 manual

data rate selection

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

RW

Y

N

GRP1_OV_CNT[14]

GRP1_OV_CNT[13]

GRP1_OV_CNT[12]

GRP1_OV_CNT[11]

GRP1_OV_CNT[10]

GRP1_OV_CNT[9]

GRP1_OV_CNT[8]

GRP0_OV_DLTA[3]

GRP0_OV_DLTA[2]

GRP0_OV_DLTA[1]

GRP0_OV_DLTA[0]

GRP1_OV_DLTA[3]

GRP1_OV_DLTA[2]

GRP1_OV_DLTA[1]

GRP1_OV_DLTA[0]

RESERVED

Group 1 count MSB

64

65

66

Sets the PPM delta tolerance for the

PPM counter lock check for group 0.

Must also program channel

Reg_0x67[7].

Sets the PPM delta tolerance for the

PPM counter lock check for group 1.

Must also program channel

Reg_0x67[6].

RESERVED

61

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

67

7

6

5

0

1

RW

Y

GRP0_OV_DLTA[4]

GRP1_OV_DLTA[4]

HV_LOCKMON_EN

1: Enable periodic monitoring of

HEO/VEO for lock qualification.

0: Disable periodic HEO/VEO

monitoring for lock qualification.

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

1

0

1

0

1

0

RW

N

Y

RESERVED

68

69

6A

6B

RESERVED

VEO_LCK_THRSH[3]

VEO_LCK_THRSH[2]

VEO_LCK_THRSH[1]

VEO_LCK_THRSH[0]

HEO_LCK_THRSH[3]

HEO_LCK_THRSH[2]

HEO_LCK_THRSH[1]

HEO_LCK_THRSH[0]

RESERVED

VEO threshold to meet before lock is

established. The LSB step size is 4

counts of VEO.

HEO threshold to meet before lock is

established. The LSB step size is 4

counts of HEO.

RESERVED

FOM_A[6]

Alternate Figure of Merit variable A.

Max value for this register is 128.

FOM_A[5]

FOM_A[4]

FOM_A[3]

FOM_A[2]

FOM_A[1]

FOM_A[0]

62

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

6C

6D

6E

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

0

RW

Y

FOM_B[7]

FOM_B[6]

FOM_B[5]

FOM_B[4]

FOM_B[3]

FOM_B[2]

FOM_B[1]

FOM_B[0]

FOM_C[7]

FOM_C[6]

FOM_C[5]

FOM_C[4]

FOM_C[3]

FOM_C[2]

FOM_C[1]

FOM_C[0]

HEO adjustment for Alternate FoM,

variable B

VEO adjustment for Alternate FoM,

variable C

EN_NEW_FOM_CTLE

1: CTLE adaption state machine will

use the alternate FoM

HEO_ALT = (HEO-B)*A*2VEO_ALT

= (VEO-C)*(1-A)*2

The values of A,B,C are set in

channel Reg_0x6B, 0x6C, and 0x6D.

The value of A is equal to the

register value divided by 128.

The Alternate FoM = (HEOB)*A*2 +

(VEO-C)*(1-A)*2

6

0

RW

Y

EN_NEW_FOM_DFE

1: DFE adaption state machine will

use the alternate FoM.

HEO_ALT = (HEO-B)*A*2VEO_ALT

= (VEO-C)*(1-A)*2

The values of A,B,C are set in

channel Reg_0x6B, 0x6C, and 0x6D.

The value of A is equal to the

register value divided by 128

The Alternate FoM = (HEOB)*A*2 +

(VEO-C)*(1-A)*2

5

4

3

2

1

0

RW

N

RESERVED

63

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

6F

7

0

RW

Y

MR_EN_LOW_DIVSEL_EQ

Normally, during adaptation, if the

divider setting is >2, then a fixed EQ

setting, from Reg_0x3A will be used.

However, if Reg_0x6F[7]=1, then an

EQ adaptation will be performed

instead.

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

0

1

0

1

0

RW

R

Y

N

Y

N

RESERVED

EQ_LB_CNT[3]

EQ_LB_CNT[2]

EQ_LB_CNT[1]

EQ_LB_CNT[0]

PRBS_INT

RESERVED

70

CTLE look-beyond count for

adaptation

71

When enabled by Reg_0x31[7], goes

HI if a PRBS stream is detected.

Clears on reading.

PRBS checker must be enabled with

Reg_0x30[3].

Once cleared, if a PRBS error

occurs, then the interrupt will again

go HI. Clears on reading.

If signal detect is lost, this is

considered a PRBS error, and the

interrupt will go HI. Clears on

reading.

6

5

4

3

2

1

0

7

6

5

4

0

R

N

RESERVED

DFE_POL_1_OBS

DFE_WT1_OBS[4]

DFE_WT1_OBS[3]

DFE_WT1_OBS[2]

DFE_WT1_OBS[1]

DFE_WT1_OBS[0]

RESERVED

DFE tap 1 polarity observation

DFE tap 1 weight observation

72

RESERVED

DFE_POL_2_OBS

Primary observation point for DFE

tap 2 polarity

3

2

1

0

R

N

DFE_WT2_OBS[3]

DFE_WT2_OBS[2]

DFE_WT2_OBS[1]

DFE_WT2_OBS[0]

Primary observation point for DFE

tap 2 weight

64

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

73

74

75

7

6

5

4

0

R

N

RESERVED

DFE_POL_3_OBS

Primary observation point for DFE

tap 3 polarity

3

2

1

0

7

6

5

4

0

R

N

DFE_WT3_OBS[3]

DFE_WT3_OBS[2]

DFE_WT3_OBS[1]

DFE_WT3_OBS[0]

RESERVED

Primary observation point for DFE

tap 3 weight

RESERVED

DFE_POL_4_OBS

Primary observation point for DFE

tap 4 polarity

3

2

1

0

7

6

5

4

0

R

N

DFE_WT4_OBS[3]

DFE_WT4_OBS[2]

DFE_WT4_OBS[1]

DFE_WT4_OBS[0]

RESERVED

Primary observation point for DFE

tap 4 weight

RESERVED

DFE_POL_5_OBS

Primary observation point for DFE

tap 5 polarity

3

2

1

0

7

6

5

4

3

2

1

0

7

0

1

0

1

0

R

N

Y

N

DFE_WT5_OBS[3]

Primary observation point for DFE

tap 5 weight

R

DFE_WT5_OBS[2]

R

DFE_WT5_OBS[1]

R

DFE_WT5_OBS[0]

76

RW

POST_LOCK_VEO_THR[3]

POST_LOCK_VEO_THR[2]

POST_LOCK_VEO_THR[1]

POST_LOCK_VEO_THR[0]

POST_LOCK_HEO_THR[3]

POST_LOCK_HEO_THR[2]

POST_LOCK_HEO_THR[1]

POST_LOCK_HEO_THR[0]

PRBS_GEN_POL_EN

VEO threshold after LOCK is

established

HEO threshold after LOCK is

established

77

1: Force polarity inversion on

generated PRBS data

6

5

4

3

2

1

0

1

0

1

0

RW

Y

N

RESERVED

65

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

78

7

6

5

0

R

N

RESERVED

SD_STATUS

Primary observation point for signal

detect status

4

3

0

R

N

CDR_LOCK_STATUS

CDR_LOCK_INT

Primary observation point for CDR

lock status

Requires that channel Reg_0x79[1]

be set.

1: Indicates CDR has achieved lock,

lock goes from LOW to HIGH. This

bit is cleared after reading. This bit

will stay set until it has been cleared

by reading.

2

0

R

N

SD_INT

Requires that channel Reg_0x79[0]

be set.

1: Indicates signal detect status has

changed. This will trigger when

signal detect goes from LOW to

HIGH or HIGH to LOW. This bit is

cleared after reading. This bit will

stay set until it has been cleared by

reading.

1

0

R

N

EOM_VRANGE_LIMIT_ERROR

HEO_VEO_INT

Goes high if GET_HEO_VEO

indicates high during adaptation

Requires that channel Reg_0x36[6]

be set.

1: Indicates that HEO/VEO dropped

below the limits set in channel

Reg_0x76 This bit is cleared after

reading. This bit will stay set until it

has been cleared by reading.

79

7

6

0

RW

N

RESERVED

PRBS_CHKR_EN

1: Enable the PRBS checker.

0: Disable the PRBS checker

5

0

RW

N

PRBS_GEN_EN

1: Enable the pattern generator

0: Disable the pattern generator

4

3

2

1

0

RW

N

Y

RESERVED

CDR_LOCK_INT_EN

1: Enable CDR lock interrupt,

observable in channel Reg_0x78[3]

0: Disable CDR lock interrupt

0

RW

Y

SD_INT_EN

1: Enable signal detect interrupt,

observable in channel Reg_0x78[3]

0: Disable signal detect interrupt

7A

7

6

5

4

3

2

1

0

RW

N

RESERVED

66

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

7B

7C

7D

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

0

RW

R

N

Y

RESERVED

PRBS_FIXED[7]

PRBS_FIXED[6]

PRBS_FIXED[5]

PRBS_FIXED[4]

PRBS_FIXED[3]

PRBS_FIXED[2]

PRBS_FIXED[1]

PRBS_FIXED[0]

Pattern generator user defined

pattern LSB. MSB located at channel

Reg_0x97.

R

RW

CONT_ADAPT_HEO_CHNG_TH Limit for HEO change before

RS[3]

triggering a DFE adaption while

continuous DFE adaption is enabled.

6

5

4

3

2

1

0

1

0

1

0

RW

Y

CONT_ADAPT_HEO_CHNG_TH

RS[2]

CONT_ADAPT_HEO_CHNG_TH

RS[1]

CONT_ADAPT_HEO_CHNG_TH

RS[0]

CONT_ADAPT_VEO_CHNG_TH Limit for VEO change before

RS[3]

triggering a DFE adaption while

continuous DFE adaption is enabled.

(Refer to the Programming Guide for

more details)

CONT_ADAPT_VEO_CHNG_TH

RS[2]

CONT_ADAPT_VEO_CHNG_TH

RS[1]

CONT_ADAPT_VEO_CHNG_TH

RS[0]

7E

7

6

5

4

3

2

1

0

1

0

1

RW

Y

CONT_ADPT_TAP_INCR[3]

CONT_ADPT_TAP_INCR[2]

CONT_ADPT_TAP_INCR[1]

CONT_ADPT_TAP_INCR[0]

RESERVED

Limit for allowable tap increase from

the previous base point

RESERVED

67

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

EN_OBS_ALT_FOM

DESCRIPTION

7F

7

0

RW

N

1: Allows for alternate FoM

calculation to be shown in channel

registers Reg_0x27, Reg_0x28 and

Reg_0x29 instead of HEO and VEO

6

5

4

0

1

0

RW

N

Y

RESERVED

EN_DFE_CONT_ADAPT

1: Continuous DFE adaption is

enabled

0: DFE adapts only during lock and

then freezes

(Refer to the Programming Guide for

more details)

3

1

RW

Y

CONT_ADPT_CMP_BOTH

1: If continuous DFE adaption is

enabled, a DFE adaption will trigger

if either HEO orVEO degrades

2

1

0

1

0

RW

Y

CONT_ADPT_COUNT[2]

CONT_ADPT_COUNT[1]

CONT_ADPT_COUNT[0]

Limit for number of weights the DFE

can look ahead in continuous

adaption.

(Refer to the Programming Guide for

more details)

80

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

1

0

R

N

RESERVED

81

68

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

82

7

0

RW

N

FREEZE_PRBS_CNTR

1: Freeze the PRBS error count to

allow for readback.

0: Normal operation. Error counters

is allowed to increment if the PRBS

checker is properly configured

6

5

0

RW

N

RST_PRBS_CNTS

PRBS_PATT_OV

1: Reset the PRBS error counter.

0: Normal operation. Error counter is

released from reset.

1: Override PRBS pattern auto-

detection. Forces the pattern checker

to only lock onto the pattern defined

in Reg_0x82[4:2].

0: Normal operation. Pattern checker

will automatically detect the PRBS

pattern

4

3

2

0

RW

N

PRBS_PATT[2]

PRBS_PATT[1]

PRBS_PATT[0]

Used with the PRBS checker. Usage

is enabled with Reg_0x82[5]. Select

PRBS pattern to be checked:

000 - PRBS7

001 - PRBS9

010 - PRBS11

011 - PRBS15

100 - PRBS23

101 - PRBS31

110 - PRBS58

111 - PRBS63

1

0

RW

N

PRBS_POL_OV

PRBS_POL

1: Override PRBS pattern auto

polarity detection. Forces the pattern

checker to only lock onto the polarity

defined in bit 0 of this register.

0: Normal operation, pattern checker

will automatically detect the PRBS

pattern polarity

Usage is enabled with

Reg_0x82[1]=1

0: Forced polarity = true

1: Forced polarity = inverted

83

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

R

N

RESERVED

PRBS_ERR_CNT[10]

PRBS_ERR_CNT[9]

PRBS_ERR_CNT[8]

PRBS_ERR_CNT[7]

PRBS_ERR_CNT[6]

PRBS_ERR_CNT[5]

PRBS_ERR_CNT[4]

PRBS_ERR_CNT[3]

PRBS_ERR_CNT[2]

PRBS_ERR_CNT[1]

PRBS_ERR_CNT[0]

PRBS checker error count

84

PRBS checker error count

69

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

85

86

87

88

89

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

R

N

RESERVED

70

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

8A

8B

8C

8D

8E

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

1

0

R

N

Y

RESERVED

R

RESERVED

VGA_SEL_GAIN

RESERVED

R

RW

VGA selection bit :

1: VGA high-gain mode

0: VGA low-gain mode

(Refer to the Programming Guide for

more details)

71

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

8F

90

91

7

6

0

R

N

EQ_BST_TO_EQ[7]

Primary observation point for the EQ

boost setting.

R

EQ_BST_TO_EQ[6]

EQ_BST_TO_EQ5]

EQ_BST_TO_EQ[4]

EQ_BST_TO_EQ[3]

EQ_BST_TO_EQ[2]

EQ_BST_TO_EQ[1]

EQ_BST_TO_EQ[0]

RESERVED

5

R

4

R

3

R

2

R

1

R

0

R

7

RW

RESERVED

6

RESERVED

5

RESERVED

4

RESERVED

3

RESERVED

2

RESERVED

1

RESERVED

0

RESERVED

7

RESERVED

6

RESERVED

5

RESERVED

4

RESERVED

3

RESERVED

2

RESERVED

1

RESERVED

0

RESERVED

92

93

94

95

7:0

7

RESERVED

SD_ENABLE

1: Force enable signal detect

0: Normal operation

6

5

0

RW

N

SD_DISABLE

1: Force disable signal detect

0: Normal operation

DC_OFF_ENABLE

1: Force enable DC offset

compensation

0: Normal operation

4

0

RW

N

DC_OFF_DISABLE

1: Force disable DC offset

compensation

0: Normal operation

3

2

1

0

RW

N

EQ_ENABLE

EQ_DISABLE

1: Force enable the CTLE

0: Normal operation

1: Force disable the CTLE

0: Normal operation

1

0

RW

N

RESERVED

72

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

96

7

6

5

4

3

0

1

RW

N

Y

RESERVED

EQ_EN_LOCAL

RESERVED

1: Enable the ebuf for the local

output. Can be set independently of

other controls.

(Refer to the Programming Guide for

more details)

2

1

0

RW

Y

EQ_EN_FANOUT

EQ_SEL_XPNT

XPNT_SLAVE

1: Enable the ebuf for the fanout.

Can be set independently of other

controls.

(Refer to the Programming Guide for

more details)

1: Indicates to a channel where it is

getting its data from. 0 indicates

local. 1-indicates from the cross.

(Refer to the Programming Guide for

more details)

1: Indicates to a channel if it needs to

wait for the other channel to

complete its lock/adaptation. The

need for this condition comes up

when input of one channel is routed

to the other channel or multiple

channels.

(Refer to the Programming Guide for

more details)

97

7

6

0

1

R

N

Y

PRBS_FIXED[15]

PRBS_FIXED[14]

PRBS_FIXED[13]

PRBS_FIXED[12]

PRBS_FIXED[11]

PRBS_FIXED[10]

PRBS_FIXED[9]

PRBS_FIXED[8]

RESERVED

Pattern generator user defined

pattern MSB. LSB located at channel

Reg_0x7C.

5

R

4

R

3

R

2

R

1

R

0

R

98

99

7:6

5:0

7

RW

RESERVED

6

RESERVED

5

RESERVED

4

RESERVED

3

RESERVED

2

RESERVED

1

RESERVED

0

RESERVED

73

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

9A

9B

9C

9D

9E

7

6

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

RW

R

Y

N

Y

N

Y

N

Y

N

RESERVED

CP_EN_IDAC_PD[2]

CP_EN_IDAC_PD[1]

CP_EN_IDAC_PD[0]

CP_EN_IDAC_FD[2]

CP_EN_IDAC_FD[1]

CP_EN_IDAC_FD[0]

RESERVED

NOT USED

RESERVED

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

6

5

4

3

2

1

0

7

Phase detector charge pump setting,

when override is enabled. See

reg_0C for other bits.

6

5

4

Frequency detector charge pump

setting, when override is enabled.

See reg_0C for other bits.

3

2

1

RESERVED

0

9F

A0

A1

A2

A3

A4

7:0

R

NOT USED

R

NOT USED

R

NOT USED

R

NOT USED

R

NOT USED

74

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Table 11. Channel Registers, 3A to A9 (continued)

DEFAULT

VALUE

(Hex)

ADDRESS

(Hex)

BITS

MODE

EEPROM

FIELD NAME

DESCRIPTION

A5

7

6

5

0

1

RW

Y

PFD_SEL_DATA_PSTLCK[2]

PFD_SEL_DATA_PSTLCK[1]

PFD_SEL_DATA_PSTLCK[0]

Output mode for when the CDR is in

lock. For these values to take effect,

Reg_0x09[5] must be set to 0, which

is the default.

000: Raw Data

001: Retimed data (default)

100: PRBS Generator or Fixed

Pattern Generator Data

101: 10M clock

111: Mute

All other values are reserved. (Refer

to the Programming Guide for more

details)

4

3

2

1

0

7

0

RW

N

RESERVED

INCR_HIST_TMR

RESERVED

A6

Provides an option to increase EOM

timer given by 0x2A[7:4] for

histogram collection by +8 for

selection values < 8

6

1

RW

Y

EOM_TMR_ABRT_ON_HIT

Enables faster scan through the eye-

matrix by moving on to the next

matrix point as soon as hit is

observed

Note: This bit does not affect when

slope measurement are in progress

5

4

0

RW

Y

SLP_MIN_REQ_HITS[1]

SLP_MIN_REQ_HITS[0]

Minimum required hit count for

registering a hit during slope

measurements.

3

0

RW

Y

LFT_SLP

0: allows slope measurement for the

right side of the eye

1: allows slope measurement for the

left side of the eye

2

0

RW

Y

TOP_SLP

0: allows slope measurement for the

bottom side of the eye

1: allows slope measurement for the

top side of the eye

1

0

1

RW

Y

DFE_BATHTUB_FOM

CTLE_BATHTUB_FOM

Enables slope-based bathtub FoM

for DFE adaptation

Enables slope-based bathtub FoM

for CTLE adaptation

A7

A8

A9

7:0

0

R

N

Y

RESERVED

RW

75

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

9 Application and Implementation

注

Information in the following applications sections is not part of the TI component

specification, and TI does not warrant its accuracy or completeness. TI’s customers are

responsible for determining suitability of components for their purposes. Customers should

validate and test their design implementation to confirm system functionality.

9.1 Application Information

The DS250DF410 is a high-speed retimer which extends the reach of differential channels and cleans jitter and

other signal impairments in the process. It can be deployed in a variety of different systems from backplanes to

front ports to active cable assemblies. The following sections outline typical applications and their associated

design considerations.

9.2 Typical Applications

The DS250DF410 is typically used in the following application scenarios:

1. Front-Port Jitter Cleaning Applications

2. Active Cable Applications

3. Backplane and Mid-plane Applications

Line Card

Switch Fabric

25G-LR

x4

25G-VSR

x4

DS250DF410

Optical

x4

25G-LR

SFP28/QSFP28

ASIC

FPGA

Active Copper

DS250DF410

x4

25G-LR

x4 25G

DS250DF410

QSFP28

Backplane/

Midplane

图 15. Typical Uses for the DS250DF410 in a System

76

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Typical Applications (接下页)

9.2.1 Front-Port Jitter Cleaning Applications

The DS250DF410 has strong equalization capabilities that allow it to equalize insertion loss, reduce jitter, and

extend the reach of front-port interfaces. A single DS250DF410 can be used to support all four egress channels

for a 100GbE port. Another DS250DF410 can be used to support all four ingress channels for the same 100GbE

ports. Alternatively, a single DS250DF410 can be used to support all egress channels for four 25GbE ports, and

another DS250DF410 can be used to support all four ingress channels for the same four 25GbE ports.

For applications which require IEEE802.3 100GBASE-CR4 or 25GBASE-CR auto-negotiation and link training, a

linear repeater device such as the DS280BR810 (or similar) is recommended.

图 16 illustrates this configuration, and 图 17 shows an example simplified schematic for a typical front-port

application.

Network Interface Card (NIC) or Host Bus Adapter (HBA)

25G/28G-LR

25G/28G-VSR

DS250DF410

Optical or

fixed-rate Copper

x4

DS250DF410

1 x 100GbE QSFP28

ASIC

PCIe

FPGA

Optical or Copper

25G/28G-LR

25G/28G-VSR

x4

Copper

Optical or Copper

or

Optical

Optical or Copper

DS250DF410

x4

DS250DF410

Optical or Copper

4 x 25GbE SFP28

or

2 x 50GbE SFP28

图 16. Front-Port Application Block Diagram

77

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

Typical Applications (接下页)

AC coupling

capacitors needed

No AC coupling

capacitors needed

Egress Retimer

RX0P

RX0N

TX0P

TX0N

CDR

RX

TX

.

RX3P

RX3N

TX3P

TX3N

CDR

RX

TX

2.5V or

3.3V

VDD

SMBus

Slave mode

To other

open-drain

interrupt pins

1 kΩ

INT_N

EN_SMB

SDA

SDC

To system

SMBus⁽¹⁾

Address straps

(pull-up, pull-

down, or float)

ADDR0

ADDR1

NC_TEST[5:0]

CAL_CLK_IN

To next device‘s

CAL_CLK_IN

25 MHz

CAL_CLK_OUT

SMBus Slave

mode

ALL_DONE_N

GND

READ_EN_N

VDD

2.5V

Minimum

recommended

decoupling

0.1ꢀF

(4x)

1ꢀF

(2x)

1 x QSFP28

Or

4 x SFP28

Host ASIC /

FPGA

AC coupling

capacitors needed

No AC coupling

capacitors needed

Ingress Retimer

TX0P

RX0P

CDR

TX

RX

TX0N

RX0N

.

TX7P

TX7N

RX7P

RX7N

CDR

TX

RX

VDD

SMBus

Slave mode

1 kΩ

INT_N

EN_SMB

SDA

SDC

Address straps

(pull-up, pull-

down, or float)

ADDR0

ADDR1

NC_TEST[5:0]

CAL_CLK_IN

CAL_CLK_OUT

SMBus Slave

mode

ALL_DONE_N

GND

READ_EN_N

VDD

2.5V

Minimum

recommended

decoupling

0.1ꢀF

(4x)

1ꢀF

(2x)

(1) SMBus signals need to be pulled up elsewhere in the system.

图 17. Front-Port Application Schematic

9.2.1.1 Design Requirements

For this design example, the following guidelines outlined in 表 12 apply.

78

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Typical Applications (接下页)

表 12. Front-Port Application Design Guidelines

DESIGN PARAMETER

REQUIREMENT

Egress (ASIC-to-module) direction: AC coupling capacitors in the

range of 100 to 220 nF are required for the RX inputs and are NOT

required for the TX outputs.

Ingress (module-to-ASIC) direction: AC coupling capacitors in the

range of 100 to 220 nF are required for the TX outputs and are NOT

required for the RX inputs.

AC coupling capacitors

Input channel insertion loss

Output channel insertion loss

≤ 35 dB at 25.78125 Gbps Nyquist frequency (12.9 GHz)

Egress (ASIC-to-module) direction: Follow CAUI-4 / CEI-25G-VSR

host channel requirements (approximately 7dB at 12.9 GHz).

Ingress (module-to-ASIC) direction: Depends on downstream ASIC /

FPGA capabilities. The DS250DF410 has a low-jitter output driver

with 3-tap FIR filter for equalizing a portion of the output channel.

Host ASIC TX launch amplitude

Host ASIC TX FIR filter

800 mVppd to 1200 mVppd.

Depends on channel loss. Refer to Setting the Output V_OD, Pre-

Cursor, and Post-Cursor Equalization.

9.2.1.2 Detailed Design Procedure

The design procedure for front-port applications is as follows:

1. Determine the total number of channels on the board which require a DS250DF410 for signal conditioning.

This will dictate the total number of DS250DF410 devices required for the board. It is generally

recommended that channels connected to the same front-port cage be grouped together in the same

DS250DF410 device. This will simplify the device settings, as similar loss channels generally utilize similar

settings.

2. Determine the maximum current draw required for all DS250DF410 retimers. This may impact the selection

of the regulator for the 2.5-V supply rail. To calculate the maximum current draw, multiply the maximum

transient power supply current by the total number of DS250DF410 devices.

3. Determine the maximum operational power consumption for the purpose of thermal analysis. There are two

ways to approach this calculation:

a. Maximum mission-mode operational power consumption is when all channels are locked and re-

transmitting the data which is received. PRBS pattern checkers/generators are not used in this mode

because normal traffic cannot be checked with a PRBS checker. For this calculation, multiply the worst-

case power consumption in mission mode by the total number of DS250DF410 devices.

b. Maximum debug-mode operational power consumption is when all channels are locked and re-

transmitting the data which is received. At the same time, some channels’ PRBS checkers or generators

may be enabled. For this calculation, multiply the worst-case power consumption in debug mode by the

total number of DS250DF410 devices.

4. Determine the SMBus address scheme needed to uniquely address each DS250DF410 device on the board,

depending on the total number of devices identified in step 2. Each DS250DF410 can be strapped with one

of 16 unique SMBus addresses. If there are more DS250DF410 devices on the board than the number of

unique SMBus addresses which can be assigned, then use an I2C expander like the TCA/PCA family of

I2C/SMBus switches and multiplexers to split up the SMBus into multiple busses.

5. Determine if the device will be configured from EEPROM (SMBus Master Mode) or from the system I2C bus

(SMBus Slave Mode).

a. If SMBus Master Mode will be used, provisions should be made for an EEPROM on the board with 8-bit

SMBus address 0xA0. Refer to SMBus Master Mode for more details on SMBus Master Mode including

EEPROM size requirements.

b. If SMBus Slave Mode will be used for all device configurations, an EEPROM is not needed.

6. Make provisions in the schematic and layout for standard decoupling capacitors between the device VDD

supply and GND. Refer to the pin function description in Pin Configuration and Functions for more details.

7. Make provisions in the schematic and layout for a 25 MHz (±100 ppm) single-ended CMOS clock. Each

DS250DF410 retimer buffers the clock on the CAL_CLK_IN pin and presents the buffered clock on the

CAL_CLK_OUT pin. This allows multiple (up to 20) retimers’ calibration clocks to be daisy chained to avoid

79

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

the need for multiple oscillators on the board. If the oscillator used on the board has a 2.5-V CMOS output,

then no AC coupling capacitor or resistor ladder is required at the input to CAL_CLK_IN. No AC coupling or

resistor ladder is needed between one retimer’s CAL_CLK_OUT output and the next retimer’s CAL_CLK_IN

input. The final retimer’s CAL_CLK_OUT output can be left floating.

8. Connect the INT_N open-drain output to an FPGA or CPU if interrupt monitoring is desired. Note that

multiple retimers’ INT_N outputs can be connected together because this is an open-drain output. The

common INT_N net should be pulled high.

9. If the application requires initial CDR lock acquisition at the ambient temperature extremes defined in

Recommended Operating Conditions, care should be taken to ensure the operating junction temperature is

met as well as the CDR stay-in-lock ambient temperature range defined in Timing Requirements, Retimer

Jitter Specifications. For example, if initial CDR lock acquisition occurs at an ambient temperature of 85 ºC,

then maintaining CDR lock would require the ambient temperature surrounding the DS250DF410 to be kept

above (85 ºC - TEMP_LOCK-).

9.2.1.3 Application Curves

图 18. DS250DF410 Operating at 25.78125 Gbps

图 19. DS250DF410 FIR Transmit Equalization While

Operating at 25.78125 Gbps

图 18 shows a typical output eye diagram for the DS250DF410 operating at 25.78125 Gbps with PRBS9 pattern

using FIR main-cursor of +18, pre-cursor of -1 and post-cursor of +2. All other device settings are left at default.

图 19 shows an example of DS250DF410 FIR transmit equalization while operating at 25.78125 Gbps. In this

example, the Tx FIR filter main-cursor is set to +15, post-cursor to -3 and pre-cursor to -3. An 8T pattern is used

to evaluate the FIR filter, which consists of 0xFF00. All other device settings are left at default.

80

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

9.2.2 Active Cable Applications

The DS250DF410 has strong equalization capabilities that allow it to recover data over long and/or thin-gauge

copper cables. A single DS250DF410 can be used on a QSFP28 paddle card to create a half-active cable

assembly which is longer and/or thinner than passive cables. Alternatively, two DS250DF410 devices can be

used on a QSFP28 paddle card to create a full-active cable assembly and achieve even longer reach and/or

thinner cables.

图 20 illustrates these configurations, 图 21 shows an example simplified schematic for a half-active cable

application, and 图 22 shows an example simplified schematic for a full-active cable application.

Line Card

Full-Active Cable

DS250DF410

QSFP

x4 25G VSR

ASIC

FPGA

Half-Active Cable

DS250DF410

QSFP

x4 25G VSR

DS250DF410

图 20. Active Cable Application Block Diagram

81

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

No Retimer and no AC coupling capacitors needed

Paddle

card host

Paddle

card cable

side

AC coupling

capacitors needed

RX Retimer

AC coupling

capacitors needed

TX0P

RX0P

CDR

TX

RX

side

TX0N

RX0N

.

TX7P

TX7N

RX7P

RX7N

CDR

TX

RX

2.5V or

3.3V

VDD

To micro-

controller

interrupt pin

SMBus

Slave mode

1 kΩ

INT_N

EN_SMB

SDA

SDC

To paddle

card SMBus

Address straps

(pull-up, pull-

down, or float)

ADDR0

ADDR1

NC_TEST[5:0]

CAL_CLK_IN

25 MHz

CAL_CLK_OUT

SMBus Slave

mode

ALL_DONE_N

GND

READ_EN_N

VDD

2.5V

Minimum

recommended

decoupling

0.1ꢀF

(4x)

1ꢀF

(2x)

图 21. Half-Active Cable Application Schematic

82

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

TX Retimer

RX0P

RX0N

TX0P

TX0N

CDR

RX

TX

.

RX3P

RX3N

TX3P

TX3N

CDR

RX

TX

2.5V or

3.3V

VDD

To micro-

controller

interrupt pin

SMBus

Slave mode

1 kΩ

INT_N

EN_SMB

SDA

SDC

To paddle

card SMBus

Address straps

(pull-up, pull-

down, or float)

ADDR0

ADDR1

NC_TEST[5:0]

CAL_CLK_IN

To next device‘s

CAL_CLK_IN

25 MHz

CAL_CLK_OUT

SMBus Slave

mode

ALL_DONE_N

GND

READ_EN_N

VDD

2.5V

Minimum

recommended

decoupling

0.1ꢀF

(4x)

1ꢀF

(2x)

Paddle

card host

side

Paddle

AC coupling

capacitors needed

RX Retimer

AC coupling

card cable

capacitors needed

TX0P

RX0P

CDR

TX

RX

side

TX0N

RX0N

.

TX7P

TX7N

RX7P

RX7N

CDR

TX

RX

VDD

1 kΩ

SMBus

Slave mode

INT_N

EN_SMB

SDA

SDC

Address straps

(pull-up, pull-

down, or float)

ADDR0

ADDR1

NC_TEST[5:0]

CAL_CLK_IN

CAL_CLK_OUT

SMBus Slave

mode

ALL_DONE_N

GND

READ_EN_N

VDD

2.5V

Minimum

recommended

decoupling

0.1ꢀF

(4x)

1ꢀF

(2x)

图 22. Full-Active Cable Application Schematic

83

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

9.2.2.1 Design Requirements

For this design example, the following guidelines outlined in 表 13 and 表 14 apply.

表 13. Half-Active Cable Application Design Guidelines

DESIGN PARAMETER

REQUIREMENT

Place the DS250DF410 on the receive side of the paddle card such

that it is receiving data from the cable, and transmitting towards the

host.

Device placement

100 nF AC coupling capacitors are required for the RX inputs and

the TX outputs.

AC coupling capacitors

The raw cable insertion loss including the insertion loss of the paddle

card should be ≤ 27 dB at 25.78125 Gbps Nyquist frequency (12.9

GHz). This is to ensure that the total loss at the input to the

DS250DF410 is ≤ 35 dB at 12.9 GHz. Assuming a worst-case host-

side PCB loss of 7 dB, plus a connector loss of 1 dB, the remaining

loss allocated for the raw cable and paddle cards is 27 dB.

Cable insertion loss

表 14. Full-Active Cable Application Design Guidelines

DESIGN PARAMETER

REQUIREMENT

A full-active QSFP cable will utilize two pieces of DS250DF410 per

paddle card. Typically, one device will be placed on each side of the

paddle card.

Device placement

Transmit-side Retimer: 100 nF AC coupling capacitors are required

for the RX inputs and are not required for the TX outputs. This link

segment will be AC coupled on the paddle card at the opposite end

of the cable.

Receive-side Retimer: 100 nF AC coupling capacitors are required

for the RX inputs and the TX outputs.

AC coupling capacitors

Cable insertion loss

The raw cable insertion loss including the insertion loss of the paddle

card should be ≤ 35 dB at 25.78125 Gbps Nyquist frequency (12.9

GHz).

9.2.2.2 Detailed Design Procedure

The design procedure for active cable applications is as follows:

1. Determine the maximum current draw required for the DS250DF410 retimer(s) on the paddle card. This may

impact the selection of the regulator for the 2.5-V supply rail. To calculate the maximum current draw,

multiply the maximum transient power supply current by the total number of DS250DF410 devices.

2. Determine the maximum operational power consumption for the purpose of thermal analysis. There are two

ways to approach this calculation:

a. Maximum mission-mode operational power consumption is when all channels are locked and re-

transmitting the data which is received. PRBS pattern checkers/generators are not used in this mode

because normal traffic cannot be checked with a PRBS checker. For this calculation, multiply the worst-

case power consumption in mission mode by the total number of DS250DF410 devices.

b. Maximum debug-mode operational power consumption is when all channels are locked and re-

transmitting the data which is received. At the same time, some channels’ PRBS checkers or generators

may be enabled. For this calculation, multiply the worst-case power consumption in debug mode by the

total number of DS250DF410 devices.

3. Determine the SMBus address for the DS250DF410 Retimer(s). If using just one Retimer for a half-active

cable, the ADDR[1:0] pins can be left floating for an 8-bit SMBus slave address of 0x44. If using a second

DS250DF410, as in the case of a full-active cable assembly, a single pull-up or pull-down resistor can be

used on one address pin. For example, with ADDR0 = Float and ADDR1 = 1 kΩ the 8-bit SMBus slave

address will be 0x34.

4. Determine if the device will be configured from EEPROM (SMBus Master Mode) or from the system I2C bus

(SMBus Slave Mode).

a. If SMBus Master Mode will be used, provisions should be made for an EEPROM on the board with 8-bit

SMBus address 0xA0. Refer to SMBus Master Mode for more details on SMBus Master Mode including

EEPROM size requirements.

84

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

b. If SMBus Slave Mode will be used for all device configurations, for example when the Retimer(s) is

configured with a microcontroller, an EEPROM is not needed.

5. Make provisions in the schematic and layout for standard decoupling capacitors between the device VDD

supply and GND. Refer to the pin function description in Pin Configuration and Functions for more details.

6. Make provisions in the schematic and layout for a 25 MHz (±100 ppm) single-ended CMOS clock. The

DS250DF410 retimer buffers the clock on the CAL_CLK_IN pin and presents the buffered clock on the

CAL_CLK_OUT pin. When using two Retimers on a paddle card, only one 25 MHz clock is required. The

CAL_CLK_OUT pin of one retimer can be connected to teh CAL_CLK_IN pin of the other retimer.

7. Connect the INT_N open-drain output to the paddle card MCU if interrupt monitoring is desired, otherwise

leave it floating. Note that multiple retimers’ INT_N outputs can be connected together because this is an

open-drain output. The common INT_N net should be pulled high.

8. If the application requires initial CDR lock acquisition at the ambient temperature extremes defined in

Recommended Operating Conditions, care should be taken to ensure the operating junction temperature is

met as well as the CDR stay-in-lock ambient temperature range defined in Timing Requirements, Retimer

Jitter Specifications. For example, if initial CDR lock acquisition occurs at an ambient temperature of 85 ºC,

then maintaining CDR lock would require the ambient temperature surrounding the DS250DF410 to be kept

above (85 ºC - TEMP_LOCK-).

9.2.2.3 Application Curves

See Application Curves in section Front-Port Jitter Cleaning Applications.

85

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

9.2.3 Backplane and Mid-plane Applications

The DS250DF410 has strong equalization capabilities that allow it to recover data over channels up to 35 dB

insertion loss. As a result, the optimum placement for the DS250DF410 in a backplane/mid-plane application is

with the higher-loss channel segment at the input and the lower-loss channel segment at the output. This

reduces the equalization burden on the downstream ASIC/FPGA, as the DS250DF410 is equalizing a majority of

the overall channel. This type of asymmetric placement is not a requirement, but when an asymmetric placement

is required due to the presence of a passive backplane or mid-plane, then this becomes the recommended

placement.

Passive Backplane/

Midplane

Switch Fabric Card

Line Card

x4 25G

ASIC

FPGA

x4 25G

ASIC

FPGA

图 23. Backplane/Mid-Plane Application Block Diagram

86

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

RX0P

TX0P

CDR

RX

TX

RX0N

.

TX0N

.

RX3P

RX3N

TX3P

TX3N

CDR

RX

TX

VDD

2.5 V or 3.3 V

SMBus

Slave mode

1 kΩ

To other

open-drain

interrupt pins

EN_SMB

INT_N

NC_TEST[5:0]

SDA

SDC

To system

SMBus⁽¹⁾

Address straps

(pull-up, pull-

down, or float)

ADDR0

ADDR1

25 MHz

CAL_CLK_IN

CAL_CLK_OUT

To next device‘s

CAL_CLK_IN

Output can float

in slave mode

SMBus Slave mode

2.5 V

READ_EN_N

VDD

ALL_DONE_N

GND

Minimum

recommended

decoupling

1 ꢀF

(2x)

0.1 ꢀF

(4x)

Backplane /

Mid-plane

Connector

ASIC / FPGA

TX0P

RX0P

CDR

TX

RX

TX0N

.

RX0N

.

TX3P

TX3N

RX3P

RX3N

CDR

TX

RX

VDD

SMBus

1 kΩ

Slave mode

EN_SMB

INT_N

SDA

SDC

NC_TEST[5:0]

Address straps

(pull-up, pull-

down, or float)

ADDR0

ADDR1

CAL_CLK_IN

CAL_CLK_OUT

ALL_DONE_N

GND

Output can float

in slave mode

SMBus Slave mode

2.5 V

READ_EN_N

VDD

1 ꢀF

(2x)

0.1 ꢀF

(4x)

Minimum

recommended

decoupling

图 24. Backplane/Mid-Plane Application Schematic

87

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

9.2.3.1 Design Requirements

For this design example, the following guidelines outlined in 表 15 apply.

表 15. Backplane/Mid-Plane Application Design Guidelines

DESIGN PARAMETER

REQUIREMENT

AC coupling capacitors in the range of 100 to 220 nF are required for

the RX inputs and TX outputs.

AC coupling capacitors

Input channel insertion loss

Output channel insertion loss

≤ 35 dB at 25.78125 Gbps Nyquist frequency (12.9 GHz)

Depends on downstream ASIC / FPGA capabilities. The

DS250DF410 has a low-jitter output driver with 3-tap FIR filter for

equalizing a portion of the output channel.

Link partner TX launch amplitude

Link partner TX FIR filter

800 mVppd to 1200 mVppd

Depends on channel loss. Refer to Setting the Output V_OD, Pre-

Cursor, and Post-Cursor Equalization.

9.2.3.2 Detailed Design Procedure

The design procedure for backplane/mid-plane applications is as follows:

1. Determine the total number of channels on the board which require a DS250DF410 for signal conditioning.

This will dictate the total number of DS250DF410 devices required for the board. It is generally

recommended that channels with similar total insertion loss on the board be grouped together in the same

DS250DF410 device. This will simplify the device settings, as similar loss channels generally utilize similar

settings.

2. Determine the maximum current draw required for all DS250DF410 retimers. This may impact the selection

of the regulator for the 2.5-V supply rail. To calculate the maximum current draw, multiply the maximum

transient power supply current by the total number of DS250DF410 devices.

3. Determine the maximum operational power consumption for the purpose of thermal analysis. There are two

ways to approach this calculation:

a. Maximum mission-mode operational power consumption is when all channels are locked and re-

transmitting the data which is received. PRBS pattern checkers/generators are not used in this mode

because normal traffic cannot be checked with a PRBS checker. For this calculation, multiply the worst-

case power consumption in mission mode by the total number of DS250DF410 devices.

b. Maximum debug-mode operational power consumption is when all channels are locked and re-

transmitting the data which is received. At the same time, some channels’ PRBS checkers or generators

may be enabled. For this calculation, multiply the worst-case power consumption in debug mode by the

total number of DS250DF410 devices.

4. Determine the SMBus address scheme needed to uniquely address each DS250DF410 device on the board,

depending on the total number of devices identified in step 2. Each DS250DF410 can be strapped with one

of 16 unique SMBus addresses. If there are more DS250DF410 devices on the board than the number of

unique SMBus addresses which can be assigned, then use an I2C expander like the TCA/PCA family of

I2C/SMBus switches and multiplexers to split up the SMBus into multiple busses.

5. Determine if the device will be configured from EEPROM (SMBus Master Mode) or from the system I2C bus

(SMBus Slave Mode).

a. If SMBus Master Mode will be used, provisions should be made for an EEPROM on the board with 8-bit

SMBus address 0xA0. Refer to SMBus Master Mode for more details on SMBus Master Mode including

EEPROM size requirements.

b. If SMBus Slave Mode will be used for all device configurations, an EEPROM is not needed.

6. Make provisions in the schematic and layout for standard decoupling capacitors between the device VDD

supply and GND. Refer to the pin function description in Pin Configuration and Functions for more details.

7. Make provisions in the schematic and layout for a 25MHz (±100 ppm) single-ended CMOS clock. Each

DS250DF410 retimer buffers the clock on the CAL_CLK_IN pin and presents the buffered clock on the

CAL_CLK_OUT pin. This allows multiple (up to 20) retimers’ calibration clocks to be daisy chained to avoid

the need for multiple oscillators on the board. If the oscillator used on the board has a 2.5-V CMOS output,

then no AC coupling capacitor or resistor ladder is required at the input to CAL_CLK_IN. No AC coupling or

resistor ladder is needed between one retimer’s CAL_CLK_OUT output and the next retimer’s CAL_CLK_IN

input. The final retimer’s CAL_CLK_OUT output can be left floating.

88

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

8. Connect the INT_N open-drain output to an FPGA or CPU if interrupt monitoring is desired. Note that

multiple retimers’ INT_N outputs can be connected together because this is an open-drain output. The

common INT_N net should be pulled high.

9. If the application requires initial CDR lock acquisition at the ambient temperature extremes defined in

Recommended Operating Conditions, care should be taken to ensure the operating junction temperature is

met as well as the CDR stay-in-lock ambient temperature range defined in Timing Requirements, Retimer

Jitter Specifications. For example, if initial CDR lock acquisition occurs at an ambient temperature of 85 ºC,

then maintaining CDR lock would require the ambient temperature surrounding the DS250DF410 to be kept

above (85 ºC - TEMP_LOCK-).

9.2.3.3 Application Curves

See Application Curves in section Front-Port Jitter Cleaning Applications.

89

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

10 Power Supply Recommendations

Follow these general guidelines when designing the power supply:

1. The power supply should be designed to provide the recommended operating conditions outlined in

Specifications in terms of DC voltage, AC noise, and start-up ramp time.

2. The maximum current draw for the DS250DF410 is provided in Specifications . This figure can be used to

calculate the maximum current the power supply must provide. Typical mission-mode current draw can be

inferred from the typical power consumption in Specifications.

3. The DS250DF410 does not require any special power supply filtering (that is, ferrite bead), provided the

recommended operating conditions are met. Only standard supply decoupling is required. Refer to Pin

Configuration and Functions for details concerning the recommended supply decoupling.

11 Layout

11.1 Layout Guidelines

The following guidelines should be followed when designing the layout:

1. Decoupling capacitors should be placed as close to the VDD pins as possible. Placing them directly

underneath the device is one option if the board design permits.

2. High-speed differential signals TXnP/TXnN and RXnP/RXnN should be tightly coupled, skew matched, and

impedance controlled.

3. Vias should be avoided when possible on the high-speed differential signals. When vias must be used, care

should be taken to minimize the via stub, either by transitioning through most/all layers, or by back drilling.

4. GND relief can be used beneath the high-speed differential signal pads to improve signal integrity by

counteracting the pad capacitance.

5. GND relief can be used beneath the AC coupling capacitor pads to improve signal integrity by counteracting

the pad capacitance.

6. GND vias should be placed directly beneath the device connecting the GND plane attached to the device to

the GND planes on other layers. This has the added benefit of improving thermal conductivity from the

device to the board

7. BGA landing pads for a 0.5-mm pitch flip-chip BGA are typically 0.3 mm in diameter (exposed). The actual

size of the copper pad will depend on whether solder-mask-defined (SMD) or non-solder-mask-defined

solder land pads are used. For more information, refer to TI’s Surface Mount Technology (SMT) and

Packaging application notes on the TI website.

8. If vias are used for the high-speed signals, ground via should be implemented adjacent to the signal via to

provide return path and isolation. For differential pair, the typical via configuration is ground-signal-signal-

ground.

9. Note that some BGA balls in the DS250DF410 pinout have been de-populated to allow for GND and VDD

vias to be placed with ≥1.0 mm via-to-via spacing.

11.2 Layout Example

The following example layout demonstrates how all signals can be escaped from the BGA array using microstrip

routing on a generic multi-layer stackup. This example layout assumes the following:

•

Trace width: 0.159 mm (6.25 mil)

Trace edge-to-edge spacing: 0.197 mm (7.75 mil)

VIA-to-VIA spacing: 0.7 mm (27.9 mil) minimum; Note: 1.0 mm VIA-to-VIA spacing is also achievable if PCB

manufacturing rules stipulate

•

No VIA-in-pad used

Note: Some TI test pins (i.e. NC_TEST[5:0]) are routed in this example layout, but in most applications these

pins can be left floating.

Note that many other escape routing options exist using different trace width and spacing combinations. The

optimum trace width and spacing will depend on the PCB material, PCB routing density, and other factors.

90

DS250DF410

www.ti.com.cn

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

Layout Example (接下页)

图 26. Layer 1 GND

图 25. Top Layer

图 27. Internal Low-Speed Signal Layers

图 28. VDD Layer

图 29. Bottom Layer

91

DS250DF410

ZHCSKE8D –MARCH 2016–REVISED OCTOBER 2019

www.ti.com.cn

12 器件和文档支持

12.1 文档支持

12.1.1 开发支持

更多相关信息，请参阅 TI 表面贴装技术 (SMT) 参考资料（位于

http://focus.ti.com/quality/docs 页面的“质量和无铅 (Pb) 数据”菜单下）。

12.2 文档支持

12.2.1 相关文档

请参阅如下相关文档：

•

《DS2x0DF810、DS250DFx10、DS250DF230 编程人员指南》 (SNLU182)

单击此处，请求访问 DS250DF410 MySecure 文件夹中的 DS2X0DFX10 IBIS-AMI 模型和编程人员指南。

12.3 接收文档更新通知

要接收文档更新通知，请导航至 ti.com. 上的器件产品文件夹。单击右上角的通知我进行注册，即可每周接收产品

信息更改摘要。有关更改的详细信息，请查看任何已修订文档中包含的修订历史记录。

12.4 支持资源

TI E2E™ support forums are an engineer's go-to source for fast, verified answers and design help — straight

from the experts. Search existing answers or ask your own question to get the quick design help you need.

Linked content is provided "AS IS" by the respective contributors. They do not constitute TI specifications and do

not necessarily reflect TI's views; see TI's Terms of Use.

12.5 商标

E2E is a trademark of Texas Instruments.

12.6 静电放电警告

ESD 可能会损坏该集成电路。德州仪器 (TI) 建议通过适当的预防措施处理所有集成电路。如果不遵守正确的处理措施和安装程序 , 可

能会损坏集成电路。

ESD 的损坏小至导致微小的性能降级 , 大至整个器件故障。精密的集成电路可能更容易受到损坏 , 这是因为非常细微的参数更改都可

能会导致器件与其发布的规格不相符。

12.7 Glossary

SLYZ022 — TI Glossary.

This glossary lists and explains terms, acronyms, and definitions.

13 机械、封装和可订购信息

以下页面包含机械、封装和可订购信息。这些信息是指定器件的最新可用数据。数据如有变更，恕不另行通知，且

不会对此文档进行修订。如需获取此数据表的浏览器版本，请查阅左侧的导航栏。

92

PACKAGE OUTLINE

ABM0101A

FCBGA - 1.03 mm max height

SCALE 2.300

PLASTIC BALL GRID ARRAY

6.1

5.9

A

B

BALL A1 CORNER

6.1

5.9

1.03 MAX

C

SEATING PLANE

0.08 C

BALL TYP

0.26

TYP

0.15

5 TYP

SYMM

(0.5) TYP

L

K

J

(0.5) TYP

H

G

F

SYMM

5

TYP

E

D

C

0.35

101X

0.25

0.15

0.05

C A

C

B

A

0.5 TYP

1

2

3

4

5

6

7

8

9

10 11

0.5 TYP

BALL A1 CORNER

4222100/B 09/2015

NOTES:

1. All linear dimensions are in millimeters. Any dimensions in parenthesis are for reference only. Dimensioning and tolerancing

per ASME Y14.5M.

2. This drawing is subject to change without notice.

www.ti.com

EXAMPLE BOARD LAYOUT

ABM0101A

FCBGA - 1.03 mm max height

PLASTIC BALL GRID ARRAY

(0.5) TYP

101X ( 0.28)

8

9

1

2

3

4

5

6

7

10

11

A

B

C

(0.5) TYP

D

E

F

G

H

J

SYMM

K

L

SYMM

LAND PATTERN EXAMPLE

SCALE:15X

0.05 MAX

0.05 MIN

METAL UNDER

SOLDER MASK

(

0.28)

METAL

(

0.28)

SOLDER MASK

OPENING

SOLDER MASK

OPENING

SOLDER MASK

DEFINED

NON-SOLDER MASK

DEFINED

(PREFERRED)

SOLDER MASK DETAILS

NOT TO SCALE

4222100/B 09/2015

NOTES: (continued)

3. Final dimensions may vary due to manufacturing tolerance considerations and also routing constraints.

For information, see Texas Instruments literature number SPRAA99 (www.ti.com/lit/spraa99).

www.ti.com

EXAMPLE STENCIL DESIGN

ABM0101A

FCBGA - 1.03 mm max height

PLASTIC BALL GRID ARRAY

(0.5) TYP

8

9

1

2

3

4

5

6

7

10

11

A

(0.5) TYP

B

C

D

E

F

G

H

J

SYMM

K

L

SYMM

SOLDER PASTE EXAMPLE

BASED ON 0.1 mm THICK STENCIL

SCALE:15X

101X ( 0.25)

PCB PAD

(R0.05) TYP

S

C

A

L

E

6

0

.

0

STENCIL DETAIL

NTS

4222100/B 09/2015

NOTES: (continued)

4. Laser cutting apertures with trapezoidal walls and rounded corners may offer better paste release.

www.ti.com

重要声明和免责声明

TI“按原样”提供技术和可靠性数据（包括数据表）、设计资源（包括参考设计）、应用或其他设计建议、网络工具、安全信息和其他资源，

不保证没有瑕疵且不做出任何明示或暗示的担保，包括但不限于对适销性、某特定用途方面的适用性或不侵犯任何第三方知识产权的暗示担

保。

这些资源可供使用 TI 产品进行设计的熟练开发人员使用。您将自行承担以下全部责任：(1) 针对您的应用选择合适的 TI 产品，(2) 设计、验

证并测试您的应用，(3) 确保您的应用满足相应标准以及任何其他功能安全、信息安全、监管或其他要求。

这些资源如有变更，恕不另行通知。TI 授权您仅可将这些资源用于研发本资源所述的 TI 产品的应用。严禁对这些资源进行其他复制或展示。

您无权使用任何其他 TI 知识产权或任何第三方知识产权。您应全额赔偿因在这些资源的使用中对 TI 及其代表造成的任何索赔、损害、成

本、损失和债务，TI 对此概不负责。

TI 提供的产品受 TI 的销售条款或 ti.com 上其他适用条款/TI 产品随附的其他适用条款的约束。TI 提供这些资源并不会扩展或以其他方式更改

TI 针对 TI 产品发布的适用的担保或担保免责声明。

TI 反对并拒绝您可能提出的任何其他或不同的条款。IMPORTANT NOTICE

邮寄地址：Texas Instruments, Post Office Box 655303, Dallas, Texas 75265


型号：	DS250DF410ABMT
厂家：	TEXAS INSTRUMENTS
描述：	25Gbps 多速率 4 通道重定时器 \| ABM \| 101 \| -40 to 85 接口集成电路
文件：	总100页 (文件大小：2761K)
中文：	中文翻译
下载：	下载PDF数据表文档文件

DS250DF410ABMT [TI]

相关型号：

DS250DF810

DS250DF810ABVR

DS250DF810ABVT

DS2540-7D0TB2

DS2540-7DDTB1

DS2540-7ED0TB2

DS2540-7EDDTB2

DS2540-7EPD0

DS2540-7EPD0B2

DS2540-7EPD0T

DS2540-7EPD0TB1

DS2540-7EPD0TB2