DP83865 [TI]
DP83865 Gig PHYTER V 10/100/1000 Ethernet Physical Layer;型号: | DP83865 |
厂家: | TEXAS INSTRUMENTS |
描述: | DP83865 Gig PHYTER V 10/100/1000 Ethernet Physical Layer 局域网(LAN)标准 |
文件: | 总88页 (文件大小:628K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
DP83865
DP83865 Gig PHYTER V 10/100/1000 Ethernet Physical Layer
Literature Number: SNLS165B
October 2004
DP83865 Gig PHYTER® V
10/100/1000 Ethernet Physical Layer
General Description
The DP83865 is a fully featured Physical Layer transceiver
with integrated PMD sublayers to support 10BASE-T,
100BASE-TX and 1000BASE-T Ethernet protocols.
■ Integrated PMD sublayer featuring adaptive equalization
and baseline wander compensation according to ANSI
X3.T12
The DP83865 is an ultra low power version of the DP83861
and DP83891. It uses advanced 0.18 um, 1.8 V CMOS
technology, fabricated at National Semiconductor’s South
Portland, Maine facility.
■ 3.3 V or 2.5 V MAC interfaces:
■ IEEE 802.3u MII
■ IEEE 802.3z GMII
The DP83865 is designed for easy implementation of
10/100/1000 Mb/s Ethernet LANs. It interfaces directly to
Twisted Pair media via an external transformer. This device
interfaces directly to the MAC layer through the IEEE
802.3u Standard Media Independent Interface (MII), the
IEEE 802.3z Gigabit Media Independent Interface (GMII),
or Reduced GMII (RGMII).
■ RGMII version 1.3
■ User programmable GMII pin ordering
■ IEEE 802.3u Auto-Negotiation and Parallel Detection
■ Fully Auto-Negotiates between 1000 Mb/s, 100 Mb/s,
The DP83865 is a fourth generation Gigabit PHY with field
proven architecture and performance. Its robust perfor-
mance ensures drop-in replacement of existing
10/100 Mbps equipment with ten to one hundred times the
performance using the existing networking infrastructure.
and 10 Mb/s full duplex and half duplex devices
■ Speed Fallback mode to achieve quality link
■ Cable length estimator
■ LED support for activity, full / half duplex, link1000,
link100 and link10, user programmable (manual on/off),
or reduced LED mode
Applications
The DP83865 fits applications in:
■ 10/100/1000 Mb/s capable node cards
■ Switches with 10/100/1000 Mb/s capable ports
■ High speed uplink ports (backbone)
■ Supports 25 MHz operation with crystal or oscillator.
■ Requires only two power supplies, 1.8 V (core and
analog) and 2.5 V (analog and I/O). 3.3V is supported
as an alternative supply for I/O voltage
Features
■ User programable interrupt
■ Ultra low power consumption typically 1.1 watt
■ Supports Auto-MDIX at 10, 100 and 1000 Mb/s
■ Supports JTAG (IEEE1149.1)
■ Fully compliant with IEEE 802.3 10BASE-T, 100BASE-
TX and 1000BASE-T specifications
■ 128-pin PQFP package (14mm x 20mm)
SYSTEM DIAGRAM
MII
GMII
RGMII
10BASE-T
100BASE-TX
1000BASE-T
DP83820
10/100/1000 Mb/s
ETHERNET MAC
DP83865
10/100/1000 Mb/s
ETHERNET PHYSICAL LAYER
25 MHz
crystal or oscillator
STATUS
LEDs
PHYTER® is a registered trademark of National Semiconductor Corporation
© 2004 National Semiconductor Corporation
www.national.com
Block Diagram
MGMT INTERFACE
COMBINED MII / GMII / RGMII INTERFACE
µC MGMT
& PHY CNTRL
MUX/DMUX
MII
GMII
100BASE-TX
Block
10BASE-T
Block
1000BASE-T
Block
MII
GMII
MII
1000BASE-T
PCS
Echo cancellation
Crosstalk cancellation
ADC
100BASE-TX
10BASE-T
PLS
PCS
Decode/Descramble
Equalization
Timing
Skew compensation
BLW
1000BASE-T
PMA
100BASE-TX
PMA
10BASE-T
PMA
Manchester
10 Mb/s
100BASE-TX
PMD
PAM-5
17 Level PR Shaped
125 Msymbols/s
MLT-3
100 Mb/s
DAC/ADC
SUBSYSTEM
TIMING
DRIVERS/
RECEIVERS
DAC/ADC
TIMING BLOCK
MAGNETICS
4-pair CAT-5 Cable
www.national.com
2
Table of Contents
1.0
Pin Description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 6.0
Electrical Specifications . . . . . . . . . . . . . . . . . . . . . . . 71
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
MAC Interfaces (MII, GMII, and RGMII) . . . . . . . 5
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
DC Electrical Specification . . . . . . . . . . . . . . . . . 71
Reset Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Clock Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
1000 Mb/s Timing . . . . . . . . . . . . . . . . . . . . . . . . 74
RGMII Timing . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
100 Mb/s Timing . . . . . . . . . . . . . . . . . . . . . . . . . 77
10 Mb/s Timing . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Loopback Timing . . . . . . . . . . . . . . . . . . . . . . . . 79
Serial Management Interface Timing . . . . . . . . . 80
Management Interface
Media Dependent Interface
. . . . . . . . . . . . . . . . . . . . 7
. . . . . . . . . . . . . . . . 7
JTAG Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Clock Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Device Configuration and LED Interface . . . . . . . . 8
Reset . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Power and Ground Pins . . . . . . . . . . . . . . . . . . . . 11
Special Connect Pins . . . . . . . . . . . . . . . . . . . . 11
1.10 Pin Assignments in the Pin Number Order . . . . 12
6.10 Power Consumption . . . . . . . . . . . . . . . . . . . . . . 81
2.0
3.0
Register Block . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 7.0
Frequently Asked Questions . . . . . . . . . . . . . . . . . . . 82
2.1
2.2
2.3
Register Definitions . . . . . . . . . . . . . . . . . . . . . . . 18
Register Map . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Register Description . . . . . . . . . . . . . . . . . . . . . . 21
7.1
7.2
7.3
7.4
7.5
Do I need to access any MDIO register to start up
the PHY? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
I am trying to access the registers through MDIO
and I got invalid data. What should I do? . . . . . 82
Why can the PHY establish a valid link but can
not transmit or receive data? . . . . . . . . . . . . . . . 82
What is the difference between TX_CLK,
TX_TCLK, and GTX_CLK? . . . . . . . . . . . . . . . . 82
What happens to the TX_CLK during 1000 Mbps
operation? Similarly what happens to RXD[4:7]
during 10/100 Mbps operation? . . . . . . . . . . . . . 82
What happens to the TX_CLK and RX_CLK
Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
Accessing Expanded Memory Space . . . . . . . . . 40
Manual Configuration . . . . . . . . . . . . . . . . . . . . . . 40
Auto-Negotiation . . . . . . . . . . . . . . . . . . . . . . . . . 41
Auto-Negotiation Register Set . . . . . . . . . . . . . . . 44
Auto-MDIX resolution . . . . . . . . . . . . . . . . . . . . . . 44
Polarity Correction . . . . . . . . . . . . . . . . . . . . . . . . 45
PHY Address, Strapping Options and LEDs . . . . 45
Reduced LED Mode . . . . . . . . . . . . . . . . . . . . . . 45
Modulate LED on Error . . . . . . . . . . . . . . . . . . . . 45
7.6
7.7
during Auto-Negotiation and during idles? . . . . . 82
Why doesn’t the Gig PHYTER V complete Auto-
Negotiation if the link partner is a forced
3.10 MAC Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.11 Clock to MAC Enable . . . . . . . . . . . . . . . . . . . . . . 46
3.12 MII/GMII/RGMII Isolate Mode . . . . . . . . . . . . . . . 46
3.13 Loopback Mode . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.14 IEEE 802.3ab Test Modes . . . . . . . . . . . . . . . . . . 46
3.15 Interrupt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.16 Low Power Mode / WOL . . . . . . . . . . . . . . . . . . . 47
3.17 Power Down Mode . . . . . . . . . . . . . . . . . . . . . . . 47
3.18 BIST Configuration . . . . . . . . . . . . . . . . . . . . . . . 47
3.19 Cable Length Indicator . . . . . . . . . . . . . . . . . . . . . 48
3.20 10BASE-T Half Duplex Loopback . . . . . . . . . . . . 48
3.21 I/O Voltage Selection . . . . . . . . . . . . . . . . . . . . . . 48
3.22 Non-compliant inter-operability mode . . . . . . . . . 48
Functional Description . . . . . . . . . . . . . . . . . . . . . . . . . 49
1000 Mbps PHY? . . . . . . . . . . . . . . . . . . . . . . . . 82
What determines Master/Slave mode when Auto-
Negotiation is disabled in 1000Base-T mode? . . 82
How long does Auto-Negotiation take? . . . . . . . 83
7.8
7.9
7.10 How do I measure FLP’s? . . . . . . . . . . . . . . . . . 83
7.11 I have forced 10 Mbps or 100 Mbps operation but
the associated speed LED doesn’t come on. . . . 83
7.12 I know I have good link, but register 0x01, bit 2
“Link Status” doesn’t contain value ‘1’ indicating
good link. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
7.13 Your reference design shows pull-up or pull-down
resistors attached to certain pins, which conflict
4.0
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
1000BASE-T PCS Transmitter . . . . . . . . . . . . . . 49
1000BASE-T PMA Transmitter . . . . . . . . . . . . . . 50
1000BASE-T PMA Receiver . . . . . . . . . . . . . . . . 50
1000BASE-T PCS Receiver . . . . . . . . . . . . . . . . 51
Gigabit MII (GMII) . . . . . . . . . . . . . . . . . . . . . . . . 52
Reduced GMII (RGMII) . . . . . . . . . . . . . . . . . . . . 53
10BASE-T and 100BASE-TX Transmitter . . . . . . 54
10BASE-T and 100BASE-TX Receiver . . . . . . . . 57
Media Independent Interface (MII) . . . . . . . . . . . . 60
with the pull-up or pull-down information specified
in the datasheet? . . . . . . . . . . . . . . . . . . . . . . . . 83
7.14 How is the maximum package case temperature
calculated? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
7.15 The DP83865 will establish Link in 100 Mbps
mode with a Broadcom part, but it will not
establish link in 1000 Mbps mode. When this
happens the DP83865’s Link LED will blink on
and off. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
7.16 How do I quickly determine the quality of the
link over the cable ? . . . . . . . . . . . . . . . . . . . . . . 83
7.17 What is the power up sequence for DP83865? . 83
7.18 What are some other applicable documents? . . 84
Physical Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . 86
5.0
Design Guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
Hardware Reset . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Clocks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Power Supply Decoupling . . . . . . . . . . . . . . . . . . 64
Sensitive Supply Pins . . . . . . . . . . . . . . . . . . . . . 64 8.0
PCB Layer Stacking . . . . . . . . . . . . . . . . . . . . . . . 64
Layout Notes on MAC Interface . . . . . . . . . . . . . . 66
Twisted Pair Interface . . . . . . . . . . . . . . . . . . . . . 66
RJ-45 Connections . . . . . . . . . . . . . . . . . . . . . . . 67
LED/Strapping Option . . . . . . . . . . . . . . . . . . . . . 67
5.10 Unused Pins and Reserved Pins . . . . . . . . . . . . . 67
5.11 I/O Voltage Considerations . . . . . . . . . . . . . . . . . 68
5.12 Power-up Recommendations . . . . . . . . . . . . . . . 68
5.13 Component Selection . . . . . . . . . . . . . . . . . . . . . 68
3
www.national.com
PQFP Pin Layout
NON_IEEE_STRAP
1
102
101
100
99
98
97
96
95
94
93
92
91
90
89
88
87
86
85
84
83
82
81
80
79
78
77
76
75
74
73
72
71
70
69
68
67
66
65
BG_REF
RESERVED
2
2V5_AVDD1
1V8_AVDD3
VSS
INTERRUPT
3
IO_VDD
4
VSS
5
1V8_AVDD2
VSS
TX_TCLK / MAN_MDIX_STRAP
6
ACTIVITY_LED / SPEED0_STRAP
7
2V5_AVDD2
PHYADDR4_STRAP
MULTI_EN_STRAP / TX_TRIGGER
VSS
LINK10_LED / RLED/SPEED1_STRAP
8
LINK100_LED / DUPLEX_STRAP
9
LINK1000_LED / AN_EN_STRAP
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
CORE_VDD
CORE_VDD
VSS
VSS
DUPLEX_LED / PHYADDR0_STRAP
IO_VDD
PHYADDR1_STRAP
IO_VDD
VSS
MDIX_EN_STRAP
MAC_CLK_EN_STRAP
CLK_OUT
CLK_IN
PHYADDR2_STRAP
PHYADDR3_STRAP
CORE_VDD
VSS
DP83865DVH
Gig PHYTER V
CLK_TO_MAC
RESERVED
IO_VDD
IO_VDD
VSS
VSS
MDC
RESERVED
TCK
MDIO
GTX_CLK/TCK
VSS
CORE_VDD
VSS
IO_VDD
TMS
TXD0/TX0
TXD1/TX1
VSS
TDO
IO_VDD
VSS
CORE_VDD
TXD2/TX2
TXD3/TX3
VSS
TDI
TRST
RESET
VDD_SEL_STRAP
CORE_VDD
VSS
IO_VDD
TXD4
TXD5
IO_VDD
VSS
TXD6
TXD7
Figure 1. DP83865 Pinout
Order Part Number: DP83865DVH
www.national.com
4
1.0 Pin Description
The DP83865 pins are classified into the following interface
categories (each is described in the sections that follow):
— MAC Interfaces
Type: I
Inputs
— Management Interface
— Media Dependent Interface
— JTAG Interface
Type: O
Output
Type: O_Z
Type: I/O_Z
Type: S
Tristate Output
Tristate Input_Output
Strapping Pin
Internal Pull-up
Internal Pull-down
— Clock Interface
— Device Configuration and LED Interface
— Reset
Type: PU
Type: PD
— Power and Ground Pins
— Special Connect Pins
1.1 MAC Interfaces (MII, GMII, and RGMII)
PQFP
Signal Name
Type
Description
Pin #
CRS/RGMII_SEL0
O_Z,
S, PD
40
CARRIER SENSE or RGMII SELECT: CRS is asserted high to indicate the
presence of a carrier due to receive or transmit activity in Half Duplex mode.
For 10BASE-T and 100BASE-TX Full Duplex operation CRS is asserted when
a received packet is detected. This signal is not defined for 1000BASE-T Full
Duplex mode.
In RGMII mode, the CRS is not used. This pin can be used as a RGMII strap-
ping selection pin.
RGMII_SEL1 RGMII_SEL0
MAC Interface
= GMII
0
0
1
1
0
1
0
1
= GMII
= RGMII - HP
= RGMII - 3COM
COL/CLK_MAC_FREQ O_Z,
39
COLLISION DETECT: Asserted high to indicate detection of a collision condi-
tion (assertion of CRS due to simultaneous transmit and receive activity) in
Half Duplex modes. This signal is not synchronous to either MII clock
(GTX_CLK, TX_CLK or RX_CLK). This signal is not defined and stays low for
Full Duplex modes.
S, PD
CLOCK TO MAC FREQUENCY Select:
1 = CLOCK TO MAC output is 125 MHz
0 = CLOCK TO MAC output is 25 MHz
TX_CLK/RGMII_SEL1 O_Z,
60
TRANSMIT CLOCK or RGMII SELECT: TX_CLK is a continuous clock signal
generated from reference CLK_IN and driven by the PHY during 10 Mbps or
100 Mbps MII mode. TX_CLK clocks the data or error out of the MAC layer and
into the PHY.
S, PD
The TX_CLK clock frequency is 2.5 MHz in 10BASE-T and 25 MHz in
100BASE-TX mode.
Note: “TX_CLK” should not be confused with the “TX_TCLK” signal.
In RGMII mode, the TX_CLK is not used. This pin can be used as a RGMII
strapping selection pin. This pin should be pulled high for RGMII interface.
5
www.national.com
1.0 Pin Description (Continued)
PQFP
Signal Name
Type
Description
Pin #
TXD0/TX0
I
76
TRANSMIT DATA: These signals carry 4B data nibbles (TXD[3:0]) during 10
Mbps and 100 Mbps MII mode, 4-bit data (TX[3:0]) in RGMII mode, and 8-bit
data (TXD[7:0]) in 1000 Mbps GMII mode. They are synchronous to the trans-
mit clocks (TX_CLK, TCK, GTX_CLK).
TXD1/TX1
TXD2/TX2
TXD3/TX3
TXD4
75
72
71
Transmit data is input to PHY. In MII or GMII mode, the transmit data is en-
abled by TX_EN. In RGMII mode, the transmit data is enabled by TXEN_ER.
68
TXD5
67
TXD6
66
TXD7
65
TX_EN/TXEN_ER
I
62
TRANSMIT ENABLE or TRANSMIT ENABLE/ERROR: In MII or GMII mode,
it is an active high input sourced from MAC layer to indicate transmission data
is available on the TXD.
In RGMII mode, it combines the transmit enable and the transmit error signals
of GMII mode using both clock edges.
GTX_CLK/TCK
TX_ER
I
I
79
61
GMII and RGMII TRANSMIT CLOCK: This continuous clock signal is sourced
from the MAC layer to the PHY. Nominal frequency is 125 MHz.
TRANSMIT ERROR: It is an active high input used in MII mode and GMII
mode forcing the PHY to transmit invalid symbols. The TX_ER signal is syn-
chronous to the transmit clocks (TX_CLK or GTX_CLK).
In MII 4B nibble mode, assertion of Transmit Error by the controller causes the
PHY to issue invalid symbols followed by Halt (H) symbols until deassertion oc-
curs.
In GMII mode, assertion causes the PHY to emit one or more code-groups that
are invalid data or delimiter in the transmitted frame.
This signal is not used in the RGMII mode.
RX_CLK
O_Z
O_Z
57
RECEIVE CLOCK: Provides the recovered receive clocks for different modes
of operation:
2.5 MHz in 10 Mbps mode.
25 MHz in 100 Mbps mode.
125 MHz in 1000 Mps GMII mode.
This pin is not used in the RGMII mode.
RXD0/RX0
RXD1/RX1
RXD2/RX2
RXD3/RX3
RXD4
56
55
52
51
50
47
46
45
41
RECEIVE DATA: These signals carry 4-bit data nibbles (RXD[3:0]) during 10
Mbps and 100 Mbps MII mode and 8-bit data bytes (RXD[7:0]) in 1000 Mbps
GMII mode. RXD is synchronous to the receive clock (RX_CLK). Receive data
is souirced from the PHY to the MAC layer.
Receive data RX[3:0] is used in RGMII mode. The data is synchronous to the
RGMII receive clock (RCK). The receive data available (RXDV_EN) indicates
valid received data to the MAC layer.
RXD5
RXD6
RXD7
RX_ER/RXDV_ER
O_Z
O_Z
RECEIVE ERROR or RECEIVE DATA AVAILABLE/ERROR: In 10 Mbps,
100 Mbps and 1000 Mbps mode this active high output indicates that the PHY
has detected a Receive Error. The RX_ER signal is synchronous with the re-
ceive clock (RX_CLK).
In RGMII mode, the receive data available and receive error is combined
(RXDV_ER) using both rising and falling edges of the receive clock (RCK).
RX_DV/RCK
44
RECEIVE DATA VALID or RECEIVE CLOCK: In MII and GMII modes, it is as-
serted high to indicate that valid data is present on the corresponding RXD[3:0]
in MII mode and RXD[7:0] in GMII mode.
In RGMII mode, this pin is the recovered receive clock (125MHz).
www.national.com
6
1.0 Pin Description (Continued)
1.2 Management Interface
PQFP
Pin #
Signal Name
Type
Description
MDC
I
81
MANAGEMENT DATA CLOCK: Synchronous clock to the MDIO serial man-
agement input/output data. This clock may be asynchronous to the MAC trans-
mit and receive clocks. The maximum clock rate is 2.5 MHz and no minimum.
MDIO
I/O
80
3
MANAGEMENT DATA I/O: Bi-directional management instruction/data signal
that may be sourced by the management station or the PHY. This pin requires
a 2kΩ pullup resistor.
INTERRUPT
O_Z,
PU
MANAGEMENT INTERRUPT: It is an active-low open drain output indicating
to the MAC layer or to a managment interface that an interrupt has requested.
The interrupt status can be read through the Interrupt Status Register. (See
section “3.15 Interrupt” on page 47.)
If used this pin requires a 2kΩ pullup resistor. This pin is to be left floating if it
is not used.
1.3 Media Dependent Interface
PQFP
PIn #
Signal Name
MDIA_P
Type
Description
I/O
108 Media Dependent Interface: Differential receive and transmit signals.
MDIA_N
MDIB_P
MDIB_N
MDIC_P
MDIC_N
MDID_P
MDID_N
109 The TP Interface connects the DP83865 to the CAT-5 cable through a single
common magnetics transformer. These differential inputs and outputs are con-
figurable to 10BASE-T, 100BASE-TX or 1000BASE-T signalling:
114
115
The DP83865 will automatically configure the driver outputs for the proper sig-
120
121
126
127
nal type as a result of either forced configuration or Auto-Negotiation. The au-
tomatic MDI / MDIX configuration allows for transmit and receive channel
configuration and polarity configuration between channels A and B, and C and
D.
NOTE: During 10/100 Mbps operation only MDIA_P, MDIA_N, MDIB_P and
MDIB_N are active. MDIA_P and MDIA_N are transmitting only and MDIB_P
and MDIB_N are receiving only. (See section “3.5 Auto-MDIX resolution” on
page 44)
1.4 JTAG Interface
PQFP
PIn #
Signal Name
Type
Description
TRST
I, PD
32
TEST RESET: IEEE 1149.1 Test Reset pin, active low reset provides for asyn-
chronous reset of the Tap Controller. This reset has no effect on the device
registers.
This pin should be pulled down through a 2kΩ resistor if not used.
TDI
I, PU
O
31
28
27
TEST DATA INPUT: IEEE 1149.1 Test Data Input pin, test data is scanned
into the device via TDI.
This pin should be left floating if not used.
TDO
TMS
TEST DATA OUTPUT: IEEE 1149.1 Test Data Output pin, the most recent
test results are scanned out of the device via TDO.
This pin should be left floating if not used.
I, PU
TEST MODE SELECT: IEEE 1149.1 Test Mode Select pin, the TMS pin se-
quences the Tap Controller (16-state FSM) to select the desired test instruc-
tion.
This pin should be left floating if not used.
7
www.national.com
1.0 Pin Description (Continued)
PQFP
Signal Name
Type
Description
PIn #
TCK
I
24
TEST CLOCK: IEEE 1149.1 Test Clock input, primary clock source for all test
logic input and output controlled by the testing entity.
This pin should be left floating if not used.
1.5 Clock Interface
Signal Name
PQFP
Pin #
Type
Description
CLK_IN
I
86
87
CLOCK INPUT: 25 MHz oscillator or crystal input (50 ppm).
CLK_OUT
O
CLOCK OUTPUT: Second terminal for 25 MHz crystal. Must be left floating if
a clock oscillator is used.
CLK_TO_MAC
O
85
CLOCK TO MAC OUTPUT: This clock output can be used to drive the clock
input of a MAC or switch device. This output is available after power-up and is
active during all modes except during hardware or software reset. Note that the
clock frequency is selectable through CLK_MAC_FREQ between 25 MHz and
125 MHz.
To disable this clock output the MAC_CLK_EN_STRAP pin has to be tied low.
1.6 Device Configuration and LED Interface
(See section “3.7 PHY Address, Strapping Options and LEDs” on page 45 and section “5.9 LED/Strapping Option” on
page 67.)
PQFP
Pin #
Signal Name
Type
Description
NON_IEEE_STRAP
I/O,
S, PD
1
NON IEEE COMPLIANT MODE ENABLE: This mode allows interoperability
with certain non IEEE compliant 1000BASE-T transceivers.
‘1’ enables IEEE compliant operation and non-compliant operation
‘0’ enables IEEE compliant operation but inhibits non-compliant operation
Note: The status of this bit is reflected in bit 10 of register 0x10. This pin also
sets the default for and can be overwritten by bit 9 of register 0x12.
MAN_MDIX_STRAP / I/O,
6
MANUAL MDIX SETTING: This pin sets the default for manual MDI/MDIX
configuration.
TX_TCLK
S, PD
‘1’ PHY is manually set to cross-over mode (MDIX)
‘0’ PHY is manually set to straight mode (MDI)
Note: The status of this bit is reflected in bit 8 of register 0x10. This pin also
sets the default for and can be overwritten by bit 14 of register 0x12.
TX_TCLK: TX_TCLK is enabled by setting bit 7 of register 0x12. It is used to
measure jitter in Test Modes 2 and 3 as described in IEEE 802.3ab specifica-
tion. TX_TCLK should not be confused with the TX_CLK signal. See Table 12
on page 29 regarding Test Mode setting. This pin should be left floating if not
used.
www.national.com
8
1.0 Pin Description (Continued)
PQFP
Signal Name
Type
Description
Pin #
ACTIVITY_LED /
SPEED0_STRAP
I/O,
S, PD
7
SPEED SELECT STRAP: These strap option pins have 2 different functions
depending on whether Auto-Negotiation is enabled or not.
Auto-Neg disabled:
Speed[1]
Speed[0]
Speed Enabled
= Reserved
1
1
0
0
1
0
1
0
= 1000BASE-T
= 100BASE-TX
= 10BASE-T
Auto-Neg enabled (Advertised capability):
Speed[1]
Speed[0]
Speed Enabled
1
1
0
0
1
0
1
0
= 1000BASE-T, 10BASE-T
= 1000BASE-T
= 1000BASE-T, 100BASE-TX
= 1000BASE-T, 100BASE-TX, 10BASE-T
Note: The status of this bit is reflected in register 0x10.12.
ACTIVITY LED: The LED output indicates the occurrence of either idle error
or packet transfer.
LINK10_LED /RLED/
SPEED1_STRAP
I/O,
S, PD
8
SPEED SELECT STRAP: The strap option pins have 2 different functions de-
pending on whether Auto-Neg is enabled or not. See SPEED0_STRAP for de-
tails.
Note: The status of this bit is reflected in register 0x10.13.
10M GOOD LINK LED: In the standard 5-LED display mode, this LED output
indicates that the PHY has established a good link at 10 Mbps.
RLED MODE: There are two reduced LED modes, the 3-in-1 and 4-in-1
modes. Each RLED mode is enabled in register 0x13.5 and 0x1A.0.
– 3-in-1: 10, 100, and 1000 Mbps good links are combined into one LED.
– 4-in-1: 3-in-1 and activity are combined.
Note: LED steady on indicates good link and flashing indicates Tx/Rx activities.
LINK100_LED /
DUPLEX_STRAP
I/O,
S, PU
9
DUPLEX MODE: This pin sets the default value for the duplex mode. ‘1’ en-
ables Full Duplex by default, ‘0’ enables Half Duplex only.
Note: The status of this bit is reflected in bit 14 of register 0x10.
100M SPEED AND GOOD LINK LED: The LED output indicates that the PHY
has established a good link at 100 Mbps.
In 100BASE-T mode, the link is established as a result of an input receive am-
plitude compliant with TP-PMD specifications which will result in internal gen-
eration of Signal Detect. LINK100_LED will assert after the internal Signal
Detect has remained asserted for a minimum of 500 µs. LINK100_LED will de-
assert immediately following the de-assertion of the internal Signal Detect.
LINK1000_LED /
AN_EN_STRAP
I/O,
S, PU
10
AUTO-NEGOTIATION ENABLE: Input to initialize Auto-Negotiation Enable
bit (register 0 bit-12).
‘1’ enables Auto-Neg and ‘0’ disables Auto-Neg.
Note: The status of this bit is reflected in bit 15 of register 0x10. This pin also
sets the default for and can be overwritten by bit 12 of register 0x00.
1000M SPEED AND GOOD LINK LED: The LED output indicates that the
PHY has established a good link at 1000 Mbps.
In 1000BASE-T mode, the link is established as a result of training, Auto-Ne-
gotiation completed, valid 1000BASE-T link established and reliable reception
of signals transmitted from a remote PHY is received.
9
www.national.com
1.0 Pin Description (Continued)
PQFP
Signal Name
Type
Description
Pin #
DUPLEX_LED /
PHYADDR0_STRAP
I/O,
S, PU
13
PHY ADDRESS [4:0]: The DP83865 provides five PHY address-sensing pins
for multiple PHY applications. The setting on these five pins provides the base
address of the PHY.
PHYADDR1_STRAP
PHYADDR2_STRAP
PHYADDR3_STRAP
PHYADDR4_STRAP
PD
PD
PD
14
17
18
The five PHYAD[4:0] bits are registered as inputs at reset with PHYADDR4 be-
ing the MSB of the 5-bit PHY address.
Note: The status of these bit is reflected in bits 4:0 of register 0x12.
DUPLEX STATUS: The LED is lit when the PHY is in Full Duplex operation
after the link is established.
PD
95
94
MULTI_EN_STRAP /
TX_TRIGGER
I/O,
S, PD
MULTIPLE NODE ENABLE: This pin determines if the PHY advertises Master
(multiple nodes) or Slave (single node) priority during 1000BASE-T Auto-Ne-
gotiation.
‘1’ Selects multiple node priority (switch or hub)
‘0’ Selects single node priority (NIC)
Note: The status of this bit is reflected in bit 5 of register 0x10.
TX_TRIGGER: This output can be enabled during the IEEE 1000BASE-T test-
modes. This signal is not required by IEEE to perform the tests, but will help to
take measurements. TX_TRIGGER is only available in test modes 1 and 4 and
provides a trigger to allow for viewing test waveforms on an oscilloscope.
MDIX_EN_STRAP
I/O,
S, PU
89
88
AUTO MDIX ENABLE: This pin controls the automatic pair swap (Auto-MDIX)
of the MDI/MDIX interface.
‘1’ enables pair swap mode
‘0’ disables the Auto-MDIX and defaults the part into the mode preset by the
MAN_MDIX_STRAP pin.
Note: The status of this bit is reflected in bit 6 of register 0x10. This pin also
sets the default for and can be overwritten by bit 15 of register 0x12.
MAC_CLK_EN_STRAP I, S,
CLOCK TO MAC ENABLE:
/ TX_SYN_CLK
PU
‘1’ CLK_TO_MAC clock output enabled
‘0’ CLK_TO_MAC disabled
Note: This status of this pin is reflected in bit 7 of register 0x10.
TX_SYN_CLK: This output can be enabled during the IEEE 1000BASE-T test-
modes. This signal is not required by IEEE to perform the tests, but will help to
take measurements. TX_SYN_CLK is only available in test modes 1 and 4.
TX_SYN_CLK = TX_TCLK / 4 in test mode 1
TX_SYN_CLK = TX_TCLK / 6 in test mode 4
VDD_SEL_STRAP
I/O, S
34
IO_VDD SELECT: This pin selects between 2.5V or 3.3V for I/O VDD .
‘1’ selects 3.3V mode
‘0’ selects 2.5V mode
This pin must either be connected directly to ground or directly to a supply volt-
age (2.5V to 3.3V).
1.7 Reset
PQFP
Pin #
Signal Name
Type
Description
RESET
I
33
RESET: The active low RESET input allows for hard-reset, soft-reset, and TRI-
STATE output reset combinations. The RESET input must be low for a mini-
mum of 150 µs.
www.national.com
10
1.0 Pin Description (Continued)
1.8 Power and Ground Pins
(See section “5.3 Power Supply Decoupling” on page 64.)
Signal Name
IO_VDD
PQFP Pin #
Description
4, 15, 21, 29, 37, 42, 53, 58, 69, 2.5V or 3.3V I/O Supply for “MAC Interfaces”, “Management
77, 83, 90
Interface”, “JTAG Interface”, “Clock Interface”, “Device Con-
figuration and LED Interface” and “Reset”.
CORE_VDD
2V5_AVDD1
2V5_AVDD2
1V8_AVDD1
1V8_AVDD2
11, 19, 25, 35, 48, 63, 73, 92
1.8V Digital Core Supply
2.5V Analog Supply
2.5V Analog Supply
1.8V Analog Supply
101
96
103, 105, 111, 117, 123
98
1.8V Analog Supply - See section “5.4 Sensitive Supply
Pins” on page 64 for low pass filter recommendation.
1V8_AVDD3
VSS
100
1.8V Analog Supply - See section “5.4 Sensitive Supply
Pins” on page 64 for low pass filter recommendation.
5, 12, 16, 20, 22, 26, 30, 36, 38, Ground
43, 49, 54, 59, 64, 70, 74, 78, 82,
91, 93, 97, 99, 104, 106, 107,
110, 112, 113, 116, 118, 119,
122, 124, 125, 128
1.9 Special Connect Pins
PQFP
Pin #
Signal Name
BG_REF
RESERVED
TYPE
Description
I
102 Internal Reference Bias: See section “5.4 Sensitive Supply Pins” on page 64
for information on how to terminate this pin.
2, 23, These pins are reserved and must be left floating.
84
11
www.national.com
1.0 Pin Description (Continued)
1.10 Pin Assignments in the Pin Number Order
Table 1.
Type
Pin #
Data Sheet Pin Name
NON_IEEE_STRAP
Connection / Comment
1
Strap Non IEEE Compliant Mode Enable: Use a 2kΩ
pull-up resistor to enable. Leave open to disable.
2
3
RESERVED
INTERRUPT
Reserved Reserved: Leave floating.
Output INTERRUPT: Connect to MAC or management
IC. This is a tri-state pin and requires an external
2kΩ pull-up resistor if the pin is used.
4
IO_VDD
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
5
6
VSS
Ground Ground: Connect to common ground plane.
TX_TCLK
Output Transmit Test Clock: See section “1.9 Special
Connect Pins” on page 11.
7
ACTIVITY_LED / SPEED0_STRAP
LINK10_LED / RLED/SPEED1_STRAP
LINK100_LED / DUPLEX_STRAP
LINK1000_LED / AN_EN_STRAP
Strap / Activity LED / SPEED0 Select: See section
Output “5.9 LED/Strapping Option” on page 67 on how
to connect this pin for speed selection and
ACTIVITY_LED function.
8
Strap / 10M Link LED / RLED / SPEED1: See section
Output “5.9 LED/Strapping Option” on page 67 on how
to connect this pin for speed selection and
LINK10_LED function.
9
Strap / 100M Link LED / Duplex Select: See section
Output “5.9 LED/Strapping Option” on page 67 on how
to connect this pin for Duplex selection and
100_LED function.
10
Strap / 1000M Link LED / Auto-Neg. Select: See sec-
Output tion “5.9 LED/Strapping Option” on page 67 on
how to connect this pin for Auto-negotiation con-
figuration and 1000_LED function.
11
12
13
CORE_VDD
Power Core VDD: (Digital) Connect to 1.8V.
VSS
Ground Ground: Connect to common ground plane.
DUPLEX_LED / PHYADDR0_STRAP
Strap / Duplex LED / PHY Address 0: See section
Output “5.9 LED/Strapping Option” on page 67 on how
to connect this pin for PHY address configura-
tion and DUPLEX_LED function.
14
15
PHYADDR1_STRAP
IO_VDD
Strap PHY Address 1: See section
“5.9 LED/Strapping Option” on page 67 on how
to connect this pin.
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
16
17
VSS
Ground Ground: Connect to common ground plane.
PHYADDR2_STRAP
Strap PHY Address 2: See section
“5.9 LED/Strapping Option” on page 67 on how
to connect this pin
18
PHYADDR3_STRAP
Strap PHY Address 3: See section
“5.9 LED/Strapping Option” on page 67 on how
to connect this pin
19
20
21
CORE_VDD
VSS
Power Core VDD: (Digital) Connect to 1.8V.
Gound Ground: Connect to common ground plane.
IO_VDD
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
22
VSS
Ground Ground: Connect to common ground plane.
www.national.com
12
1.0 Pin Description (Continued)
Table 1.
Type
Reserved Reserved: Leave floating.
Input JTAG Test Clock: This pin should be left float-
Pin #
23
Data Sheet Pin Name
Connection / Comment
RESERVED
TCK
24
ing if not used.
25
26
27
CORE_VDD
VSS
Power Core VDD: (Digital) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
TMS
Input
JTAG Test Mode Select: This pin should be left
floating if not used.
28
29
TDO
Output JTAG Test Data Output: This pin should be left
floating if not used.
IO_VDD
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
30
31
VSS
TDI
Ground Ground: Connect to common ground plane.
Input
Input
Input
JTAG Test Data Input: This pin should be left
floating if not used.
32
TRST
JTAG Test Reset: This pin should be pulled
down through a 2kΩ resistor if not used.
33
34
RESET
Reset: Connect to board reset signal.
VDD_SEL_STRAP
Strap I/O VDD Select: Pull high to select 3.3V or low
to select 2.5V. The pin must be connected direct-
ly to power or ground (no pull-up/down resistor!).
35
36
37
CORE_VDD
VSS
Power Core VDD: (Digital) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
IO_VDD
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
38
39
VSS
COL
Ground Ground: Connect to common ground plane.
Output Collision: Connect to MAC chip through a single
50 Ω impedance trace. This output is capable of
driving 35 pF load and is not intended to drive
connectors, cables, backplanes or multiple trac-
es. This applies if the part is in 100 Mbps mode
or 1000 Mbps mode.
40
41
42
CRS/RGMII_SEL0
RX_ER/RXDV_ER
IO_VDD
Output Carrier Sense: Connect to MAC chip through a
single 50Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
Output Receive Error: Connect to MAC chip through a
single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
43
44
VSS
Ground Ground: Connect to common ground plane.
RX_DV/RCK
Output Receive Data Valid: Connect to MAC chip
through a single 50 Ω impedance trace. This out-
put is capable of driving 35 pf load and is not in-
tended to drive connectors, cables, backplanes
or multiple traces. This applies if the part is in
100 Mbps mode or 1000 Mbps mode.
13
www.national.com
1.0 Pin Description (Continued)
Table 1.
Type
Pin #
Data Sheet Pin Name
Connection / Comment
45
RXD7
RXD6
RXD5
Output Receive Data 7: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
46
47
Output Receive Data 6: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
Output Receive Data 5: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
48
49
50
CORE_VDD
VSS
Power Core VDD: (Digital) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
RXD4
Output Receive Data 4: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
51
52
53
RXD3/RX3
RXD2/RX2
IO_VDD
Output Receive Data 3: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
Output Receive Data 2: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
54
55
VSS
Ground Ground: Connect to common ground plane.
RXD1/RX1
Output Receive Data 1: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
56
RXD0/RX0
Output Receive Data 0: Connect to MAC chip through
a single 50 Ω impedance trace. This output is ca-
pable of driving 35 pf load and is not intended to
drive connectors, cables, backplanes or multiple
traces. This applies if the part is in 100 Mbps
mode or 1000 Mbps mode.
www.national.com
14
1.0 Pin Description (Continued)
Table 1.
Type
Pin #
Data Sheet Pin Name
Connection / Comment
57
RX_CLK
IO_VDD
Output Receive Clock/ Receive Byte Clock 1: Con-
nect to MAC chip through a single 50 Ω imped-
ance trace. This output is capable of driving 35
pf load and is not intended to drive connectors,
cables, backplanes or multiple traces. This ap-
plies if the part is in 100 Mbps mode or 1000
Mbps mode.
58
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
59
60
VSS
Ground Ground: Connect to common ground plane.
TX_CLK/RGMII_SEL1
Output Transmit Clock: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF.
61
62
TX_ER
Input
Transmit Error: Connect to MAC chip through a
single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF.
TX_EN/TXEN_ER
Input
Transmit Enable: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF.
63
64
65
CORE_VDD
VSS
Power Core VDD: (Digital) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
TXD7
Input
Input
Input
Input
Transmit Data 7: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF.
66
67
68
69
TXD6
Transmit Data 6: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF
TXD5
Transmit Data 5: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF
TXD4
Transmit Data 4: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF
IO_VDD
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
70
71
VSS
Ground Ground: Connect to common ground plane.
TXD3/TX3
Input
Transmit Data 3: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF
72
TXD2/TX2
Input
Transmit Data 2: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF
73
74
75
CORE_VDD
VSS
Power Core VDD: (Digital) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
TXD1/TX1
Input
Transmit Data 1: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF
76
TXD0/TX0
Input
Transmit Data 0: Connect to MAC chip through
a single 50 Ω impedance trace. This input has a
typical input capacitance of 6 pF
77
78
IO_VDD
VSS
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
Ground Ground: Connect to common ground plane.
15
www.national.com
1.0 Pin Description (Continued)
Table 1.
Type
Input
Pin #
Data Sheet Pin Name
Connection / Comment
79
GTX_CLK/TCK
GMII Transmit Clock: Connect to MAC chip
through a single 50 Ω impedance trace. This in-
put has a typical input capacitance of 6 pF
80
81
MDIO
MDC
Input / Management Data I/O: This pin requires a 2kΩ
Output parallel termination resistor (pull-up to VDD).
Input
Management Data Clock: Connect to MAC or
controller using a 50 Ω impedance trace.
82
83
VSS
Ground Ground: Connect to common ground plane.
IO_VDD
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
84
85
RESERVED
Reserved Reserved: Leave floating.
CLK_TO_MAC
Output Clock to MAC: Connect to the reference clock
input of a GMAC. Use pin
MAC_CLK_EN_STRAP to disable this function.
86
87
CLK_IN
Input
Input
Clock Input: Connect to external 25MHz refer-
ence clock source. If a crystal is used connect to
first terminal of crystal.
CLK_OUT
Clock Output: Connect to the second terminal
of a crystal. Leave floating if an external clock
source is used.
88
89
90
MAC_CLK_EN_STRAP
MDIX_EN_STRAP
IO_VDD
Strap Clock to MAC Enable: Use a 2kΩ pull-down re-
sistor to disable. Leave open to enable.
Strap Automatic MDIX Enable: Use a 2kΩ pull-down
resistor to disable. Leave open to enable.
Power I/O VDD: (Digital) Connect to 2.5V or 3.3V. The
VDD_SEL pin must be tied accordingly.
91
92
93
94
VSS
Ground Ground: Connect to common ground plane.
Power Core VDD: (Digital) Connect to 1.8V.
CORE_VDD
VSS
Ground Ground: Connect to common ground plane.
MULTI_EN_STRAP
Strap Multiple Node Enable: Use a 2kΩ pull-up resis-
tor to enable. Leave open to disable.
95
PHYADDR4_STRAP
Strap PHY Address 4: See section
“5.9 LED/Strapping Option” on page 67 on how
to connect this pin.
96
97
98
AFE_VDD
VSS
Power AFE VDD: (Analog) Connect to 2.5V.
Ground Ground: Connect to common ground plane.
PGM_VDD
Power PGM VDD: Connect to 1.8V through a low pass
filter. See section “5.4 Sensitive Supply Pins” on
page 64 for details.
99
VSS
Ground Ground: Connect to common ground plane.
100
1V8_AVDD3
Power Analog Supply: Connect to 1.8V through a low
pass filter. See section “5.4 Sensitive Supply
Pins” on page 64 for details.
101
102
BG_VDD
BG_REF
Power BG VDD: (Analog) Connect to 2.5V.
Input
BG Reference: See section “5.4 Sensitive Sup-
ply Pins” on page 64 on how to connect this pin.
103
104
105
106
RX_VDD
VSS
Power Receive VDD: (Analog) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
Power Receive VDD: (Analog) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
RX_VDD
VSS
www.national.com
16
1.0 Pin Description (Continued)
Table 1.
Type
Pin #
107
Data Sheet Pin Name
Connection / Comment
VSS
Ground Ground: Connect to common ground plane.
108
MDIA_P
Input / MDI Channel A Positive: Connect to TD+ of
Output channel A of the magnetics.
109
MDIA_N
Input / MDI Channel A Negative: Connect to TD- of
Output channel A of the magnetics.
110
111
112
113
114
VSS
Ground Ground: Connect to common ground plane.
Power Receive VDD: (Analog) Connect to 1.8 Volt.
Ground Ground: Connect to common ground plane.
Ground Ground: Connect to common ground plane.
RX_VDD
VSS
VSS
MDIB_P
Input / MDI Channel B Positive: Connect to TD+ of
Output channel B of the magnetics.
115
MDIB_N
Input / MDI Channel B Negative: Connect to TD- of
Output channel B of the magnetics.
116
117
118
119
120
VSS
Ground Ground: Connect to common ground plane.
Power Receive VDD: (Analog) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
Ground Ground: Connect to common ground plane.
RX_VDD
VSS
VSS
MDIC_P
Input / MDI Channel C Positive: Connect to TD+ of
Output channel C of the magnetics.
121
MDIC_N
Input / MDI Channel C Negative: Connect to TD- of
Output channel C of the magnetics.
122
123
124
125
126
VSS
Ground Ground: Connect to common ground plane.
Power Receive VDD: (Analog) Connect to 1.8V.
Ground Ground: Connect to common ground plane.
Ground Ground: Connect to common ground plane.
RX_VDD
VSS
VSS
MDID_P
Input / MDI Channel D Positive: Connect to TD+ of
Output channel D of the magnetics.
127
128
MDID_N
VSS
Input / MDI Channel D Negative: Connect to TD- of
Output channel D of the magnetics.
Ground Ground: Connect to common ground plane.
17
www.national.com
2.0 Register Block
2.1 Register Definitions
Register maps and address definitions are given in the following table:
Table 2. Register Block - DP83865 Register Map
Offset
Decimal
Access
Tag
Description
Basic Mode Control Register
Hex
0x00
0
1
RW
RO
RO
RO
RW
RW
RW
RW
RW
RW
RO
RO
RO
RO
RO
RW
RW
RO
RW
RO
RW
RW
RW
RW
RO
RW
RW
RW
BMCR
BMSR
0x01
Basic Mode Status Register
PHY Identifier Register #1
PHY Identifier Register #2
Auto-Negotiation Advertisement Register
Auto-Negotiation Link Partner Ability Register
Auto-Negotiation Expansion Register
Auto-Negotiation Next Page TX
Auto-Negotiation Next Page RX
1000BASE-T Control Register
1000BASE-T Status Register
Reserved
0x02
2
PHYIDR1
PHYIDR2
ANAR
0x03
3
0x04
4
0x05
5
ANLPAR
0x06
6
ANER
0x07
7
ANNPTR
0x08
8
ANNPRR
1KTCR
0x09
9
0x0A
0x0B-0x0E
0x0F
0x10
10
11-14
15
16
17
18
19
20
21
22
23
24
25
26
27-28
29
30
31
1KSTSR
Reserved
1KSCR
1000BASE-T Extended Status Register
Strap Options Register
STRAP_REG
LINK_AN
AUX_CTRL
LED_CTRL
INT_STATUS
INT_MASK
EXP_MEM_CTL
INT_CLEAR
BIST_CNT
BIST_CFG1
BIST_CFG2
Reserved
0x11
Link and Auto-Negotiation Status Register
Auxiliary Control Register
0x12
0x13
LED Control Register
0x14
Interrupt Status Register
0x15
Interrupt Mask Register
0x16
Expanded Memory Access Control
Interrupt Clear Register
0x17
0x18
BIST Counter Register
0x19
BIST Configuration Register #1
BIST Configuration Register #2
Reserved
0x1A
0x1B-0x1C
0x1D
0x1E
0x1F
EXP_MEM_DATA Expanded Memory Data
EXP_MEM_ADDR Expanded Memory Address
PHY_SUP
PHY Support Register
www.national.com
18
2.0 Register Block (Continued)
19
www.national.com
2.0 Register Block (Continued)
www.national.com
20
2.0 Register Block (Continued)
2.3 Register Description
In the register description under the ‘Default’ heading, the following definitions hold true:
— RW
— RO
— LH
— LL
— SC
— P
= Read Write access
= Read Only access
= Latched High until read, based upon the occurrence of the corresponding event
= Latched Low until read, based upon the occurrence of the corresponding event
= Register sets on event occurrence (or is manually set) and Self-Clears when event ends
= Register bit is Permanently set to a default value
— STRAP[x] = Default value read from Strapped value at device pin at Reset, where x may take the values:
[0] internal pull-down
[1] internal pull-up
Table 3. Basic Mode Control Register (BMCR) address 0x00
Bit
Bit Name
Default
Description
15
Reset
0, RW, SC
Reset:
1 = Initiate software Reset / Reset in Process.
0 = Normal operation.
This bit sets the status and control registers of the PHY to their
default states. This bit, which is self-clearing, returns a value of
one until the reset process is complete (approximately 1.2 ms for
reset duration). Reset is finished once the Auto-Negotiation pro-
cess has begun or the device has entered it’s forced mode.
14
Loopback
0, RW
Loopback:
1 = Loopback enabled.
0 = Normal operation.
The loopback function enables MII/GMII transmit data to be rout-
ed to the MII/GMII receive data path. The data loops around at
the DAC/ADC Subsystem (see block diagram page 2), bypassing
the Drivers/Receivers block. This exercises most of the PHY’s in-
ternal logic.
13
Speed[0]
STRAP[0], RW Speed Select:
When Auto-Negotiation is disabled, bits 6 and 13 select device
speed selection per table below:
Speed[1]
Speed[0]
Speed Enabled
= Reserved
= 1000 Mbps
= 100 Mbps
= 10 Mbps
1
1
0
0
1
0
1
0
(The default value of this bit is = to the strap value of pin 7 during
reset/power-on IF Auto-Negotiation is disabled.)
12
AN_EN
STRAP[1], RW Auto-Negotiation Enable:
1 = Auto-Negotiation Enabled - bits 6, 8 and 13 of this register are
ignored when this bit is set.
0 = Auto-Negotiation Disabled - bits 6, 8 and 13 determine the link
speed and mode.
(The default value of this bit is = to the strap value of pin 10 during
reset/power-on.)
21
www.national.com
2.0 Register Block (Continued)
Table 3. Basic Mode Control Register (BMCR) address 0x00
Bit
Bit Name
Default
Description
11
Power_Down
0, RW
Power Down:
1 = Power down (only Management Interface and logic active.)
0 = Normal operation.
Note: This mode is internally the same as isolate mode (bit 10).
Isolate:
10
Isolate
0, RW
1 = Isolates the Port from the MII/GMII with the exception of the
serial management. When this bit is asserted, the DP83865 does
not respond to TXD[7:0], TX_EN, and TX_ER inputs, and it pre-
sents a high impedance on TX_CLK, RX_CLK, RX_DV, RX_ER,
RXD[7:0], COL and CRS outputs.
0 = Normal operation.
9
Restart_AN
0, RW, SC
Restart Auto-Negotiation:
1 = Restart Auto-Negotiation. Re-initiates the Auto-Negotiation
process. If Auto-Negotiation is disabled (bit 12 = 0), this bit is ig-
nored. This bit is self-clearing and will return a value of 1 until
Auto-Negotiation is initiated, whereupon it will self-clear. Opera-
tion of the Auto-Negotiation process is not affected by the man-
agement entity clearing this bit.
0 = Normal operation.
8
7
Duplex
STRAP[1], RW Duplex Mode:
1 = Full Duplex operation. Duplex selection is allowed only when
Auto-Negotiation is disabled (AN_EN = 0).
0 = Half Duplex operation.
(The default value of this bit is = to the strap value of pin 9 during
reset/power-on IF Auto-Negotiation is disabled.)
Collision Test
0, RW
Collision Test:
1 = Collision test enabled.
0 = Normal operation.
When set, this bit will cause the COL signal to be asserted in re-
sponse to the assertion of TX_EN withinTBD-bit times. The COL
signal will be de-asserted within 4-bit times in response to the de-
assertion of TX_EN.
6
Speed[1]
Reserved
STRAP[0], RW Speed Select: See description for bit 13.
(The default value of this bit is = to the strap value of pin 8 during
reset/power-on IF Auto-Negotiation is disabled.)
5:0
0, RO
Reserved by IEEE: Write ignored, read as 0.
Table 4. Basic Mode Status Register (BMSR) address 0x01
15
100BASE-T4
0, P
100BASE-T4 Capable:
0 = Device not able to perform 100BASE-T4 mode.
DP83865 does not support 100BASE-T4 mode and bit should al-
ways be read back as “0”.
14
13
100BASE-X
Full Duplex
1, P
1, P
100BASE-X Full Duplex Capable:
1 = Device able to perform 100BASE-X in Full Duplex mode.
100BASE-X Half Duplex Capable:
100BASE-X
Half Duplex
1 = Device able to perform 100BASE-X in Half Duplex mode.
www.national.com
22
2.0 Register Block (Continued)
Table 4. Basic Mode Status Register (BMSR) address 0x01
12
11
10
10BASE-T
Full Duplex
1, P
1, P
0, P
10BASE-T Full Duplex Capable:
1 = Device able to perform 10BASE-T in Full Duplex mode.
10BASE-T Half Duplex Capable:
10BASE-T
Half Duplex
1 = Device able to perform 10BASE-T in Half Duplex mode.
100BASE-T2 Full Duplex Capable:
100BASE-T2
Full Duplex
0 = Device unable to perform 100BASE-T2 Full Duplex mode.
DP83865 does not support 100BASE-T2 mode and bit should al-
ways be read back as “0”.
9
100BASE-T2
Half Duplex
0, P
100BASE-T2 Half Duplex Capable:
0 = Device unable to perform 100BASE-T2 Half Duplex mode.
DP83865 does not support 100BASE-T2 mode and bit should al-
ways be read back as “0”.
8
1000BASE-T
Extended Status
1, P
1000BASE-T Extended Status Register:
1 = Device supports Extended Status Register 0x0F.
Reserved by IEEE: Write ignored, read as 0.
Preamble suppression Capable:
7
6
Reserved
0, RO
1, P
Preamble
Suppression
1 = Device able to perform management transaction with pream-
ble suppressed, 32-bits of preamble needed only once after re-
set, invalid opcode or invalid turnaround.
5
4
Auto-Negotiation
Complete
0, RO
Auto-Negotiation Complete:
1 = Auto-Negotiation process complete, and contents of registers
5, 6, 7, & 8 are valid.
0 = Auto-Negotiation process not complete.
Remote Fault
0, RO, LH
Remote Fault:
1 = Remote Fault condition detected (cleared on read or by re-
set). Fault criteria: Far End Fault Indication or notification from
Link Partner of Remote Fault.
0 = No remote fault condition detected.
Auto Configuration Ability:
3
2
Auto-Negotiation
Ability
1, P
1 = Device is able to perform Auto-Negotiation.
Link Lost Since Last Read Status:
Link Status
0, RO, LL
1 = Link was good since last read of this register. (10/100/1000
Mbps operation).
0 = Link was lost since last read of this register.
The occurrence of a link failure condition will causes the Link Sta-
tus bit to clear. Once cleared, this bit may only be set by estab-
lishing a good link condition and a read via the management
interface.
This bit doesn’t indicate the link status, but rather if the link was
lost since last read. For actual link status, either this register
should be read twice, or register 0x11 bit 2 should be read.
1
0
Jabber Detect
0, RO, LH
1, P
Jabber Detect: Set to 1 if 10BASE-T Jabber detected locally.
1 = Jabber condition detected.
0 = No Jabber.
Extended Capability
Extended Capability:
1 = Extended register capable.
23
www.national.com
2.0 Register Block (Continued)
Table 5. PHY Identifier Register #1 (PHYIDR1) address 0x02
Bit
Bit Name
Default
Description
15:0
OUI[3:18]
16’b<0010_0000 OUI Bits 3:18:
_0000_0000>, P
Bits 3 to 18 of the OUI (0x080017h) are stored in bits 15 to 0 of
this register. The most significant two bits of the OUI are ignored
(the IEEE standard refers to these as bits 1 and 2).
The PHY Identifier Registers #1 and #2 together form a unique identifier for the DP83865. The Identifier consists of a con-
catenation of the Organizationally Unique Identifier (OUI), the vendor’s model number and the model revision number. A
PHY may return a value of zero in each of the 32 bits of the PHY Identifier if desired. The PHY Identifier is intended to sup-
port network management. National’s IEEE assigned OUI is 0x080017h.
Table 6. PHY Identifier Register #2 (PHYIDR2) address 0x03
Bit
Bit Name
Default
Description
15:10
OUI[19:24]
6’b<01_0111>, P OUI Bits 19:24:
Bits 19 to 24 of the OUI (0x080017h) are mapped to bits 15 to 10
of this register respectively.
9:4
3:0
VNDR_MDL[5:0]
MDL_REV[3:0]
6’b <00_0111>, Vendor Model Number:
P
The six bits of vendor model number are mapped to bits 9 to 4
(most significant bit to bit 9).
4’b <1010>, P Model Revision Number:
Four bits of the vendor model revision number are mapped to bits
3 to 0 (most significant bit to bit 3). This field will be incremented
for all major device changes.
Table 7. Auto-Negotiation Advertisement Register (ANAR) address 0x04
Bit
Bit Name
Default
Description
Next Page Indication:
15
NP
0, RW
1 = Next Page Transfer desired.
0 = Next Page Transfer not desired.
Reserved by IEEE: Writes ignored, Read as 0.
Remote Fault:
14
13
Reserved
RF
0, RO
0, RW
1 = Advertises that this device has detected a Remote Fault.
0 = No Remote Fault detected.
12
11
Reserved
0, RO
0, RW
Reserved for Future IEEE use: Write as 0, Read as 0.
Asymmetrical PAUSE:
ASY_PAUSE
1 = MAC/Controller supports Asymmetrical Pause direction.
0 = MAC/Controller does not support Asymmetrical Pause direc-
tion.
10
9
PAUSE
0, RW
0, RO
PAUSE:
1 = MAC/Controller supports Pause frames.
0 = MAC/Controller does not support Pause frames.
100BASE-T4 Support:
100BASE-T4
1 = 100BASE-T4 supported.
0 = No support for 100BASE-T4.
DP83865 does not support 100BASE-T4 mode and this bit
should always be read back as “0”.
www.national.com
24
2.0 Register Block (Continued)
Table 7. Auto-Negotiation Advertisement Register (ANAR) address 0x04
Bit
Bit Name
Default
STRAP[1], RW 100BASE-TX Full Duplex Support:
1 = 100BASE-TX Full Duplex is supported by the local device.
Description
8
100BASE-TX
Full Duplex
0 = 100BASE-TX Full Duplex not supported.
The default value of this bit is determined by the combination of
the Duplex Enable and Speed[1:0] strap pins during reset/power-
on IF Auto-Negotiation is enabled.
The advertised speed is determined by the Speed[1:0]:
Speed[1]
Speed[0]
Speeds Enabled
0
0
1
1
0
1
0
1
= 1000B-T, 100B-TX, 10B-T
= 1000B-T, 100B-TX
= 1000B-T
= 1000B-T, 10B-T
The advertised duplex mode is determined by Duplex Mode:
0 = Half Duplex
1 = Full Duplex
7
6
100BASE-TX
(Half Duplex)
STRAP[1], RW 100BASE-TX (Half Duplex) Support:
1 = 100BASE-TX (Half Duplex) is supported by the local device.
0 = 100BASE-TX (Half Duplex) not supported.
(The default value of this bit is determined by the combination of
the Duplex Enable and Speed[1:0] strap pins during reset/power-
on IF Auto-Negotiation is enabled. See bit 8 for details.)
10BASE-T
Full Duplex
STRAP[1], RW 10BASE-T Full Duplex Support:
1 = 10BASE-T Full Duplex is supported.
0 = 10BASE-T Full Duplex is not supported.
(The default value of this bit is determined by the combination of
the Duplex Enable and Speed[1:0] strap pins during reset/power-
on IF Auto-Negotiation is enabled. See bit 8 for details.)
5
10BASE-T
STRAP[1], RW 10BASE-T (Half Duplex) Support:
(Half Duplex)
1 = 10BASE-T (Half Duplex) is supported by the local device.
0 = 10BASE-T (Half Duplex) is not supported.
(The default value of this bit is determined by the combination of
the Duplex Enable and Speed[1:0] strap pins during reset/power-
on IF Auto-Negotiation is enabled. See bit 8 for details.)
4:0
PSB[4:0]
5’b<0_0001>, P Protocol Selection Bits:
These bits contain the binary encoded protocol selector support-
ed by this port. <00001> indicates that this device supports IEEE
802.3.
This register contains the advertised abilities of this device as they will be transmitted to its link partner during Auto-Negoti-
ation.
25
www.national.com
2.0 Register Block (Continued)
Table 8. Auto-Negotiation Link Partner Ability Register (ANLPAR) address 0x05
Bit
Bit Name
Default
Description
15
NP
0, RO
Next Page Indication:
0 = Link Partner does not desire Next Page Transfer.
1 = Link Partner desires Next Page Transfer.
Acknowledge:
14
ACK
0, RO
1 = Link Partner acknowledges reception of the ability data word.
0 = Not acknowledged.
The Device’s Auto-Negotiation state machine will automatically
control this bit based on the incoming FLP bursts. Software
should not attempt to write to this bit.
13
RF
0, RO
Remote Fault:
1 = Remote Fault indicated by Link Partner.
0 = No Remote Fault indicated by Link Partner.
Reserved for Future IEEE use: Write as 0, read as 0.
Asymmetrical PAUSE:
12
11
Reserved
0, RO
0, RO
ASY_PAUSE
1 = Link Partner supports Asymmetrical Pause direction.
0 = Link Partner does not support Asymmetrical Pause direction.
PAUSE:
10
9
PAUSE
0, RO
0, RO
0, RO
0, RO
0, RO
0, RO
1 = Link Partner supports Pause frames.
0 = Link Partner does not support Pause frames.
100BASE-T4 Support:
100BASE-T4
1 = 100BASE-T4 is supported by the Link Partner.
0 = 100BASE-T4 not supported by the Link Partner.
100BASE-TX Full Duplex Support:
8
100BASE-TX
Full Duplex
1 = 100BASE-TX Full Duplex is supported by the Link Partner.
0 = 100BASE-TX Full Duplex not supported by the Link Partner.
100BASE-TX (Half Duplex) Support:
7
100BASE-TX
(Half Duplex)
1 = 100BASE-TX (Half Duplex) is supported by the Link Partner.
0 = 100BASE-TX (Half Duplex) not supported by the Link Partner.
10BASE-T Full Duplex Support:
6
10BASE-T
Full Duplex
1 = 10BASE-T Full Duplex is supported by the Link Partner.
0 = 10BASE-T Full Duplex not supported by the Link Partner.
10BASE-T (Half Duplex) Support:
5
10BASE-T
(Half Duplex)
1 = 10BASE-T (Half Duplex) is supported by the Link Partner.
0 = 10BASE-T (Half Duplex) not supported by the Link Partner.
4:0
PSB[4:0]
5’b<0_0000>, Protocol Selection Bits:
RO
Link Partners’s binary encoded protocol selector.
This register contains the advertised abilities of the Link Partner as received during Auto-Negotiation
www.national.com
26
2.0 Register Block (Continued)
Table 9. Auto-Negotiate Expansion Register (ANER) address 0x06
Bit
15:5
4
Bit Name
Reserved
PDF
Default
0, RO
Description
Reserved by IEEE: Writes ignored, Read as 0.
Parallel Detection Fault:
0, RO, LH
1 = A fault has been detected via the Parallel Detection function.
0 = A fault has not been detected via the Parallel Detection func-
tion.
3
LP_NP Able
0, RO
Link Partner Next Page Able:
1 = Link Partner does support Next Page.
0 = Link Partner supports Next Page negotiation.
Next Page Able:
2
1
NP Able
1, RO
1 = Indicates local device is able to send additional “Next Pages”.
Link Code Word Page Received:
PAGE_RX
0, RO, LH
1 =Link Code Word has been received, cleared on read of this
register.
0 = Link Code Word has not been received.
Link Partner Auto-Negotiation Able:
0
LP_AN Able
0, RO
1 = Indicates that the Link Partner supports Auto-Negotiation.
0 = Indicates that the Link Partner does not support Auto-Negoti-
ation.
This register contains additional Local Device and Link Partner status information.
Table 10. Auto-Negotiation Next Page Transmit Register (ANNPTR) address 0x07
Bit
Bit Name
Default
Description
15
NP
1, RW
Next Page Indication:
1 = Another Next Page desired.
0 = No other Next Page Transfer desired.
Acknowledge:
14
13
12
ACK
MP
0, RO
1, RW
0, RW
1 = Acknowledge of 3 consecutive FLPs.
0 = No Link Code Word received.
Message Page:
1 = Message Page.
0 = Unformatted Page.
ACK2
Acknowledge2:
1 = Will comply with message.
0 = Cannot comply with message.
Acknowledge2 is used by the next page function to indicate that
Local Device has the ability to comply with the message received.
27
www.national.com
2.0 Register Block (Continued)
Table 10. Auto-Negotiation Next Page Transmit Register (ANNPTR) address 0x07
Bit
Bit Name
Default
Description
11
TOG_TX
0, RO
Toggle:
1 = Value of toggle bit in previously transmitted Link Code Word
was logic 0.
0 = Value of toggle bit in previously transmitted Link Code Word
was logic 1.
Toggle is used by the Arbitration function within Auto-Negotiation
to ensure synchronization with the Link Partner during Next Page
exchange. This bit shall always take the opposite value of the
Toggle bit in the previously exchanged Link Code Word.
10:0
CODE[10:0]
11’b<000_0000_ This field represents the code field of the next page transmission.
1000>, RW
If the MP bit is set (bit 13 of this register), then the code shall be
interpreted as a "Message Page”, as defined in annex 28C of
IEEE 802.3u. Otherwise, the code shall be interpreted as an "Un-
formatted Page”, and the interpretation is application specific.
The default value of the CODE represents a Null Page as defined
in Annex 28C of IEEE 802.3u.
This register contains the next page information sent by this device to its Link Partner during Auto-Negotiation.
Table 11. Auto-Negotiation Next Page Receive Register (ANNPRR) address 0x08
Bit
Bit Name
Default
Description
15
NP
0, RO
Next Page Indication:
1 = Another Next Page desired.
0 = No other Next Page Transfer desired.
Acknowledge:
14
13
12
ACK
MP
0, RO
0, RO
0, RO
1 = Link Partner acknowledges reception of the next page.
0 = Not acknowledged.
Message Page:
1 = Message Page.
0 = Unformatted Page.
ACK2
Acknowledge2:
1 = Link Partner will comply with message.
0 = Cannot comply with message.
Acknowledge2 is used by the next page function to indicate that
the Link Partner has the ability to comply with the message re-
ceived.
www.national.com
28
2.0 Register Block (Continued)
Table 11. Auto-Negotiation Next Page Receive Register (ANNPRR) address 0x08
Bit
Bit Name
Default
Description
11
TOG_RX
0, RO
Toggle:
1 = Value of toggle bit in previously transmitted Link Code Word
was logic 0.
0 = Value of toggle bit in previously transmitted Link Code Word
was logic 1.
Toggle is used by the Arbitration function within Auto-Negotiation
to ensure synchronization with the Link Partner during Next Page
exchange. This bit shall always take the opposite value of the
Toggle bit in the previously exchanged Link Code Word.
10:0
CODE[10:0]
11’b<000_0
This field represents the code field of the next page transmission.
000_0000>, RO If the MP bit is set (bit 13 of this register), then the code shall be
interpreted as a "Message Page”, as defined in annex 28C of
IEEE 802.3u. Otherwise, the code shall be interpreted as an "Un-
formatted Page”, and the interpretation is application specific.
The default value of the CODE represents a Reserved for future
use as defined in Annex 28C of IEEE 802.3u.
This register contains the next page information sent by its Link Partner during Auto-Negotiation.
Table 12. 1000BASE-T Control Register (1KTCR) address 0x09
Bit
Bit Name
Default
Description
15:13
Test Mode
0, RW
Test Mode Select:
bit 15 bit 14 bit 13
Test Mode Selected
1
0
0
0
0
0
1
1
0
0
0
1
0
1
0
= Test Mode 4
= Test mode 3
= Test Mode 2
= Test Mode 1
= Normal Operation
See IEEE 802.3ab section 40.6.1.1.2 “Test modes” for more in-
formation. Output for TX_TCLK when in Test Mode is on pin 6.
12
11
10
Master / Slave
Manual Config.
Enable
0, RW
0, RW
Enable Manual Master / Slave Configuration:
1 = Enable Manual Master/Slave Configuration control.
0 = Disable Manual Master/Slave Configuration control.
Using the manual configuration feature may prevent the PHY
from establishing link in 1000Base-T mode if a conflict with the
link partner’s setting exists.
Master / Slave
Config. Value
Manual Master / Slave Configuration Value:
1 = Set PHY as MASTER when register 09h bit 12 = 1.
0 = Set PHY as SLAVE when register 09h bit 12 = 1.
Using the manual configuration feature may prevent the PHY
from establishing link in 1000Base-T mode if a conflict with the
link partner’s setting exists.
Repeater / DTE
STRAP[0], RW Advertise Device Type: Multi or single port
1 = Repeater or Switch.
0 = DTE.
(The default value of this bit is = to the strap value of pin 94 during
reset/power-on IF Auto-Negotiation is enabled.)
29
www.national.com
2.0 Register Block (Continued)
Table 12. 1000BASE-T Control Register (1KTCR) address 0x09
Bit
Bit Name
Default
Description
9
1000BASE-T
Full Duplex
STRAP[1], RW Advertise 1000BASE-T Full Duplex Capable:
1 = Advertise DTE as 1000BASE-T Full Duplex Capable.
0 = Advertise DTE as not 1000BASE-T Full Duplex Capable.
(The default value of this bit is determined by the combination of
the Duplex Enable and Speed[1:0] strap pins during reset/power-
on IF Auto-Negotiation is enabled. See register 0x04 bit 8 for de-
tails.)
8
1000BASE-T
Half Duplex
STRAP[1], RW Advertise 1000BASE-T Half Duplex Capable:
1 = Advertise DTE as 1000BASE-T Half Duplex Capable.
0 = Advertise DTE as not 1000BASE-T Half Duplex Capable.
(The default value of this bit is determined by the combination of
the Duplex Enable and Speed[1:0] strap pins during reset/power-
on IF Auto-Negotiation is enabled. See register 0x04 bit 8 for de-
tails.)
7:0
Reserved
0, RW
Reserved by IEEE: Writes ignored, Read as 0.
Table 13. 1000BASE-T Status Register (1KSTSR) address 0x0A (10’d)
Bit
Bit Name
Default
Description
15
Master / Slave
Manual Config. Fault
0, RO, LH, SC MASTER / SLAVE manual configuration fault detected:
1 = MASTER/SLAVE manual configuration fault detected.
0 = No MASTER/SLAVE manual configuration fault detected.
14
13
12
11
10
Master / Slave
Config. Resolution
0, RO
0, RO
0, RO
0, RO
0, RO
MASTER / SLAVE Configuration Results:
1 = Configuration resolved to MASTER.
0 = Configuration resolved to SLAVE.
Local Receiver Status:
Local Receiver
Status
1 = OK.
0 = Not OK.
Remote Receiver
Status
Remote Receiver Status:
1 = OK.
0 = Not OK.
LP 1000BASE-T
Full Duplex
Link Partner 1000BASE-T Full Duplex:
1 = Link Partner capable of 1000BASE-T Full Duplex.
0 = Link Partner not capable of 1000BASE-T Full Duplex.
Link Partner 1000BASE-T Half Duplex:
1 = Link Partner capable of 1000BASE-T Half Duplex.
0 = Link Partner not capable of 1000BASE-T Half Duplex.
Reserved by IEEE: Write ignored, read as 0.
IDLE Error Count
LP 1000BASE-T
Half Duplex
9:8
7:0
Reserved
0, RO
IDLE ErrorCount[7:0]
0, RO, SC
This register provides status for 1000BASE-T link.
Note: Registers 0x0B - 0x0E are Reserved by IEEE.
www.national.com
30
2.0 Register Block (Continued)
Table 14. 1000BASE-T Extended Status Register (1KSCR) address 0x0F (15’d)
Bit
Bit Name
Default
Description
1000BASE-X Full Duplex Support:
1 = 1000BASE-X is supported by the local device.
0 = 1000BASE-X is not supported.
15
1000BASE-X
Full Duplex
0, P
DP83865 does not support 1000BASE-X and bit should always
be read back as “0”.
14
1000BASE-X
Half Duplex
0, P
1000BASE-X Half Duplex Support:
1 = 1000BASE-X is supported by the local device.
0 =1000BASE-X is not supported.
DP83865 does not support 1000BASE-X and bit should always
be read back as “0”.
13
12
1000BASE-T
Full Duplex
1, P
1, P
1000BASE-T Full Duplex Support:
1 = 1000BASE-T is supported by the local device.
0 =1000BASE-T is not supported.
1000BASE-T
Half Duplex
1000BASE-T Half Duplex Support:
1 = 1000BASE-T is supported by the local device.
0 =1000BASE-T is not supported.
11:0
Reserved
0, RO
Reserved by IEEE: Write ignored, read as 0.
Table 15. Strap Option Register (STRAP_REG) address 0x10 (16’d)
Bit
Bit Name
Default
Description
15
AN Enable
STRAP[1], RO Auto-Negotiation Enable: Pin 10. Default value for bit 12 of reg-
ister 0x00.
14
Duplex Mode
Speed[1:0]
STRAP[1], RO Duplex Mode: Pin 9. Default value for bit 8 of register 0x00.
13:12
STRAP[00], RO Speed Select: Pins 8 and 7. Default value for bits 6 and 13 of reg-
ister 0x00.
11
10
Reserved
0, RO
Write as 0, ignore on read.
NC Mode Enable
STRAP[0], RO Non-Compliant Mode: Pin 1. Default value for bit 9 of register
0x12.
9
8
7
6
Reserved
Reserved
0, RO
0, RO
Write as 0, ignore on read.
Write as 0, ignore on read.
MAC Clock Enable
MDIX Enable
STRAP[1], RO MAC Clock Output Enable: Pin 88.
STRAP[1], RO Auto MDIX Enable: Pin 89. Default value for bit 15 of register
0x12.
5
Multi Enable
STRAP[0], RO Multi Port Enable: Pin 94. Default value for bit 10 of register
0x09.
4:0
PHYADDR[4:0]
STRAP[0_0001], PHY Address: Pins 95, 18, 17,14, 13. Default for bits 4:0 of reg-
RO
ister 0x1F.
This register summarizes all the strap options. These can only be changed through restrapping and resetting the PHY.
31
www.national.com
2.0 Register Block (Continued)
Table 16. Link and Auto-Negotiation Status Register (LINK_AN) address 0x11 (17’d)
Bit
Bit Name
Default
Description
15:12
TP Polarity[3:0]
0, RO
Twisted Pair Polarity Status: Indicates a polaritiy reversal on
pairs A to D ([15:12]). The PHY automatically detects this condi-
tion and adjusts for it.
1 = polarity reversed
0 = normal operation
11
10
Reserved
(Power Down Status)
MDIX Status
0, RO
0, RO
Write as 0, ignore on read.
This bit is set to indicate that the PHY is in power down mode.
MDIX Status: Indicates whether the PHY’s MDI is in straight or
cross-over mode.
1 = Cross-over mode
0 = Straight mode
9
FIFO Error
Reserved
0, RO
Transmit FIFO Error: Indicates whether a FIFO overflow or un-
derrun has occured. This bit is cleared every time link is lost.
1 = FIFO error occured
0 = normal operation
8
7
0, RO
0, RO
Write as 0, ignore on read.
Shallow Loopback
Status
Shallow Loopback Status: (As set by bit 5, register 0x12)
1 = The PHY operates in shallow loopback mode
0 = Normal operation
6
5
Deep Loopback
Status
0, RO
0, RO
Deep Loopack Status: (As set by bit 14, register 0x00)
1 = The PHY operates in deep loopback mode
0 = Normal operation
Non-Compliant
Mode Status
Non-compliant Mode Status:
‘1’ detects only in non-compliant mode
‘0’ detects in both IEEE compliant and non-compliant mode
4:3
Speed[1:0] Status STRAP[00], RO Speed Resolved: These two bits indicate the speed of operation
as determined by Auto-negotiation or as set by manual configu-
ration.
Speed[1]
Speed[0]
Speed of operation
= 1000 Mbps
= 100 Mbps
1
0
1
0
1
0
0
1
= 10 Mbps
= reserved
2
1
0
Link Status
0, RO
0, RO
0, RO
Link status:
1 = indicates that a good link is established
0 = indicates no link.
Duplex Status
Duplex status:
1 = indicates that the current mode of operation is full duplex
0 = indicates that the current mode of operation is half duplex
Master / Slave Configuration Status:
1 = PHY is currently in Master mode
0 = PHY is currently in Slave mode
Master / Slave
Config. Status
www.national.com
32
2.0 Register Block (Continued)
Table 17. Auxiliary Control Register (AUX_CTRL) address 0x12 (18’d)
Bit
Bit Name
Default
Description
15
Auto-MDIX Enable
STRAP[1], RW Automatic MDIX: Indicates (sets) whether the PHY’s capability
to automatically detect swapped cable pairs is used.
1 = Automatic MDIX mode, bit 14 is ignored
0 = Manual MDIX mode
Note: This bit is ignored and the setting of bit 14 applies if Auto-
Negotiation is disabled (AN_EN = 0). Bit 10 of register 0x11
should always be checked for the actual status of MDI/MDIX op-
eration.
14
Manual MDIX Value STRAP[0], RW Manual MDIX Value: If Manual MDIX mode is selected (Auto-
MDIX selection is disabled, bit 15 = 0) this bit sets the MDIX
mode of operation. If the PHY is in Auto-MDIX mode this bit has
no effect.
1 = cross-over mode (channels A and B are swapped)
0 = straight mode
Note: Bit 10 of register 0x11 should always be checked for the ac-
tual status of MDI/MDIX operation.
13:12
RGMII_EN[1:0]
STRAP[0]
RGMII ENABLE: These two bits enables RGMII mode or
MII/GMII mode.
RGMII_EN[1:0]
11 = RGMII - 3COM mode
10 = RGMII - HP mode
01 = GMII mode
00 = GMII mode
11:10
9
Reserved
0, RO
Write as 0, ignore on read.
Non-Compliant Mode STRAP[0], RW Non-Compliant Mode Enable: This bit enables the PHY to work
in non-IEEE compliant mode. This allows interoperabilty with cer-
tain non-IEEE compliant 1000BASE-T tranceivers.
1 = enables IEEE compliant operation and non-compliant opera-
tion
0 = enables IEEE compliant operation but inhibits non-compliant
operation
8
RGMII InBand
Status Enable
0, RW
RGMII InBand Status Enable:
1 = RGMII InBand Status enabled.
0 = RGMII InBand Status disabled.
When InBand Status is enabled, PHY places link status, speed,
and duplex mode information on RXD[3:0] between the data
frames. The InBand Status may ease the MAC layer design.
Note that this bit has no impact if bit 13 = 0.
7
6
TX_TCLK Enable
0, RW
0, RW
TX_TCLK Enable: This bit enables the TX_TCLK (pin 6) output
during the IEEE 1000BASE-T test modes.
1 = TX_TCLK ouput enabled during IEEE test modes
0 = No TX_TCLK output (default)
TX_Trigger_Syn
Enable
TX_TRIGGER and TX_SYNC Enable: This bit enables the
TX_SYNC_CLK (pin 88) and TX_TRIGGER (pin 94) output dur-
ing the IEEE 1000BASE-T modes. These signals are not required
by IEEE to perform the tests, but will help to take measurements.
0 = No signal output
1 = Signal are output during IEEE test modes
Note: TX_SYN_CLK and TX_TRIGGER are only available in test
mode 1 and 4
TX_SYN_CLK = TX_TCLK / 4 in test mode 1
TX_SYN_CLK = TX_TCLK / 6 in test mode 4
33
www.national.com
2.0 Register Block (Continued)
Table 17. Auxiliary Control Register (AUX_CTRL) address 0x12 (18’d)
Bit
Bit Name
Default
Description
5
Shallow Deep Loop-
back
0, RW
Shallow Deep Loopack Enable: (Loopback status bit 7, register
0x11)
Enable
This bit places PHY in the MAC side loopback mode. Any packet
entering into TX side appears on the RX pins immediately. This
operation bypasses all internal logic and packet does not appear
on the MDI interface.
1 = The PHY operates in shallow deep loopback mode
0 = Normal operation
4
X_Mac
0, RW
Reverse GMII Data Bit Order:
Setting this bit will reverse the pins of the TXD and RXD on the
GMII interface, respectively.
1 = TXD[7:0]=>TXD[0:7], RXD[7:0]=>RXD[0:7]
0 = Normal operation
3:1
0
Reserved
0, RO
0, RW
Write as 0, ignore on read.
Jabber Disable
Jabber Disable: (Only in 10BASE-T mode) If this bit is set the
PHY ignores all jabber conditions.
1 = disable jabber function
0 = normal operation
Table 18. LED Control Register (LED_CTRL) address 0x13 (19’d)
Bit
Bit Name
Default
Description
15:14
Activity LED
0, RW
Activity LED: This LED is active when the PHY is transmitting
data, receiving data, or detecting idle error.
The following modes are available for the ACT LED:
00 = Register controlled 0x13.3:0
01 = Forced off
10 = Blink mode (blink rate approx. 750 ms)
11 = Forced on
Note: Only in normal mode (00) LEDs reflect the actual status of
the PHY. All other modes force the driver to a permanent on, off
or blinking state.
13:12
Link10 LED
0, RW
10BASE-T Link LED: This LED is active when the PHY is linked
in 10BASE-T mode.
The following modes are available for LEDs:
00 = Normal (default)
01 = Forced off
10 = Blink mode (blink rate approx. 750 ms)
11 = Forced on
Note: Only in normal mode (00) LEDs reflect the actual status of
the PHY. All other modes force the driver to a permanent on, off
or blinking state.
11:10
9:8
Link100 LED
Link1000 LED
Duplex LED
0, RW
0, RW
0, RW
100BASE-TX Link LED: This LED is active when the PHY is
linked in 100BASE-TX mode. See Activity LED for other settings.
1000BASE-T Link LED: This LED is active when the PHY is
linked in 1000BASE-T mode. See Activity LED for other settings.
7:6
Duplex LED: This LED is active when the PHY has established
a link in Full Duplex mode. See Activity LED for other settings.
www.national.com
34
2.0 Register Block (Continued)
Table 18. LED Control Register (LED_CTRL) address 0x13 (19’d)
Bit
Bit Name
Default
Description
5
reduced LED enable
0, RW
Reduced LED Mode Enable: This bit enables the reduced LED
(RLED) mode that is different from the normal five-LED mode. In
the RLED Mode, 10M Link LED is changed to link LED or Link
and activity combined LED.
When reg 0x13.5 is enabled:
Reg 0x1A.0 = 1 - 10M Link LED displays 10/100/1000 Link
Reg 0x1A.0 = 0 - 10M LED displays 10/100/1000 Link and ACT
Note: In Link mode, the LED is steady on. In Link/ACT mode,
LED is steady on when link is achieved, and LED blinks when
there is link and activity.
4
3
2
1
0
led_on_crc
led_on_ie
0, RW
0, RW
0, RW
0, RW
0, RW
an_fallback_an
an_fallback_crc
an_fallback_ie
Table 19. Interrupt Status Register (INT_STATUS) address 0x14 (20’d)
Bit
15
14
13
Bit Name
spd_cng_int
lnk_cng_int
dplx_cng_int
Default
0, RO
0, RO
0, RO
Description
Speed Change: Asserted when the speed of a link changes.
Link Change: Asserted when a link is established or broken.
Duplex Change: Asserted when the duplex mode of a link
changes.
12
11
10
9
mdix_cng_int
pol_cng_int
0, RO
0, RO
0, RO
0, RO
0, RO
0, RO
MDIX Change: Asserted when the MDIX status changes, i.e. a
pair swap occured.
Polarity Change: Asserted when the polarity of any channel
changes.
prl_det_flt_int
mas_sla_err_int
no_hcd_int
Parallel Detection Fault: Asserted when a parallel detectin fault
has been detected.
Master / Slave Error: Asserted when the Master / Slave config-
uration in 1000BASE-T mode could not be resolved.
8
No HCD: Asserted when Auto-Negotiation could not determine a
Highest Common Denominator.
7
no_lnk_int
No Link after Auto-Negotiation: Asserted when Auto-Negotia-
tion has been completed successfully and no link could be estab-
lished.
6
5
jabber_cng_int
nxt_pg_rcvd_int
an_cmpl_int
0, RO
0, RO
0, RO
0, RO
0, RO
Jabber Change: Asserted in 10BASE-T mode when a Jabber
condition has occured or has been cleared.
Next Page Received: Asserted when a Next Page has been re-
ceived.
4
Auto-negotiation complete: Asserted when Auto-Negotiation
has been completed.
3
rem_flt_cng_int
Reserved
Remote Fault Change: Asserted when the remote fault status
changes.
2:0
Write as 0, ignore on read.
35
www.national.com
2.0 Register Block (Continued)
Table 20. Interrupt Mask Register (INT_MASK) address 0x15 (21’d)
Bit
Bit Name
Default
Description
15
spd_cng_int_msk
0, RW
Setting this bit activates the spd_cng_int interrupt. The interrupt
is masked if the bit is cleared.
14
13
12
11
10
9
lnk_cng_int_msk
dplx_cng_int_msk
mdix_cng_int_msk
pol_cng_int_msk
prl_det_flt_int_msk
mas_sla_err_int_msk
no_hcd_int_msk
no_lnk_int_msk
0, RW
0, RW
0, RW
0, RW
0, RW
0, RW
0, RW
0, RW
0, RW
0, RW
0, RW
0, RW
0, RO
Setting this bit activates the lnk_cng_int interrupt. The interrupt is
masked if the bit is cleared.
Setting this bit activates the dplx_cng_int interrupt. The interrupt
is masked if the bit is cleared.
Setting this bit activates the mdix_cng_int interrupt. The interrupt
is masked if the bit is cleared.
Setting this bit activates the pol_cng_int interrupt. The interrupt is
masked if the bit is cleared.
Setting this bit activates the prl_det_flt_int interrupt. The interrupt
is masked if the bit is cleared.
Setting this bit activates the mas_sla_err_int interrupt. The inter-
rupt is masked if the bit is cleared.
8
Setting this bit activates the no_hcd_int interrupt. The interrupt is
masked if the bit is cleared.
7
Setting this bit activates the no_lnk_int interrupt. The interrupt is
masked if the bit is cleared.
6
jabber_cng_int_msk
nxt_pg_rcvd_int_msk
an_cmpl_int_msk
rem_flt_cng_int_msk
Reserved
Setting this bit activates the jabber_cng_int interrupt. The inter-
rupt is masked if the bit is cleared.
5
Setting this bit activates the nxt_pg_rcvd_int interrupt. The inter-
rupt is masked if the bit is cleared.
4
Setting this bit activates the an_cmpl_int interrupt. The interrupt
is masked if the bit is cleared.
3
Setting this bit activates the rem_flt_cng_int interrupt. The inter-
rupt is masked if the bit is cleared.
2:0
Write as 0, ignore on read.
Table 21. Expanded Memory Access Control (Exp_mem_ctl) address 0x16 (22’d)
Bit
Bit Name
Default
Description
15
Global Reset
0, RW, SC
Global Reset:
This bit resets the entire chip.
Write as 0, ignore on read.
14:8
7
Reserved
0, RO
0, RW
Broadcast Enable
Broadcast Enable:
1 = Respond to broadcast write at MDIO address 0
0 = Respond to MDIO address set in register 0x1F.4:0
Write as 0, ignore on read.
6:2
1:0
Reserved
0, RO
Address Control
[11], RW
Address Control:
00 = 8-bit expanded memory read/write (auto-incr disabled)
01 = 8-bit expanded memory read/write (auto-incr enabled)
10 = 16-bit expanded memory read/write (auto-incr enabled)
11 = 8-bit expanded memory write-only (auto-incr disabled)
www.national.com
36
2.0 Register Block (Continued)
Table 22. Interrupt Clear Register (INT_CLEAR) address 0x17 (23’d)
Bit
15
14
13
12
11
10
9
Bit Name
spd_cng_int_clr
lnk_cng_int_clr
dplx_cng_int_clr
mdix_cng_int_clr
pol_cng_int_clr
prl_det_flt_int_clr
mas_sla_err_int_clr
no_hcd_int_clr
Default
Description
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RW, SC
0, RO
Setting this bit clears the spd_cng_int interrupt.
Setting this bit clears the lnk_cng_int interrupt.
Setting this bit clears the dplx_cng_int interrupt.
Setting this bit clears the mdix_cng_int interrupt.
Setting this bit clears the pol_cng_int interrupt.
Setting this bit clears the prl_det_flt_int interrupt.
Setting this bit clears the mas_sla_err_int interrupt.
Setting this bit clears the no_hcd_int interrupt.
Setting this bit clears the no_lnk_int interrupt.
Setting this bit clears the jabber_cng_int interrupt.
Setting this bit clears the nxt_pg_rcvd_int interrupt.
Setting this bit clears the an_cmpl_int interrupt.
Setting this bit clears the rem_flt_cng_int interrupt.
Write as 0, ignore on read.
8
7
no_lnk_int_clr
6
jabber_cng_int_clr
nxt_pg_rcvd_int_clr
an_cmpl_int_clr
rem_flt_cng_int_clr
Reserved
5
4
3
2:0
Table 23. BIST Counter Register (BIST_CNT) address 0x18 (24’d)
Bit
Bit Name
Default
Description
15:0
BIST Counter
0, RO
BIST Counter: This register counts receive packets or receive
errors according to bit 15 in register BIST_CFG1. It shows either
the upper or lower 16 bit of a 32 bit value which can be selected
through bit 14 in register BIST_CFG2.
Table 24. BIST Configuration Register 1 (BIST_CFG1) address 0x19 (25’d)
Bit
Bit Name
Default
Description
15
bist_cnt_type
0, RW
Set BIST Counter Type:
1 = BIST_CNT counts receive CRC errors
0 = BIST_CNT counts receive packets
14
13
bist_cnt_clr
0, RW, SC
0, RW
BIST Counter Clear: Setting this bit clears the BIST_CNT regis-
ter to 0.
tx_bist_pak_len
Transmit BIST Packet Length:
1 = 1514 bytes
0 = 60 bytes
12
11
tx_bist_ifg
tx_bist_en
0, RW
Transmit BIST Interframe Gap: This bit sets the IFG for transmit
BIST packets.
1 = 9.6 us
0 = 0.096us
0, RW, SC
Transmit BIST Enable: This bit starts the transmit BIST. The
number of selected packets or a continous data stream is sent out
when set. This bit self-clears after the packets have been sent.
1 = Transmit BIST enabled
0 = Transmit BIST disabled
37
www.national.com
2.0 Register Block (Continued)
Table 24. BIST Configuration Register 1 (BIST_CFG1) address 0x19 (25’d)
Bit
Bit Name
Default
Description
Transmit BIST Packet Type:
10
tx_bist_pak_type
0, RW
1 = PSR9
0 = User defined packet
Write as 0, ignore on read.
9:8
7:0
Reserved
0, RO
0, RW
tx_bist_pak
User Defined Packet Content: This field sets the packet content
for the transmit BIST packets if the user defined packet type in bit
10 is selected.
Table 25. BIST Configuration Register 2 (BIST_CFG2) address 0x1A (26’d)
Bit
Bit Name
Default
Description
15
rx_bist_en
0, RW
Receive BIST Enable: This bit enables the receive BIST
counter. The BIST counter operation does not interfere with nor-
mal PHY operation.
0 = BIST counter disabled
1 = BIST counter enabled
14
bist_cnt_sel
0, RW
0, RW
BIST Counter Select: This bit selects whether the upper or lower
16 bit of the 32 bit counter value are shown in the BIST_CNT reg-
ister.
0 = displays lower 16 bit
1 = displays upper 16 bit
13:11
tx_bist_pak_cnt
Transmit BIST Packet Count: Sets the number of transmit pack-
ets
000 = continuous transmit
001 = 1 packet
010 = 10 packets
011 = 100 packets
100 = 1,000 packets
101 = 10,000 packets
110 = 100,000 packets
111 = 10,000,000 packets
10:1
0
Reserved
0, RO
0, RW
Write as 0, ignore on read.
Link/Link-ACT sel
Link/Link-ACT Select: This bit has no impact when Reg 0x13.5
= 0.
1 = LINK only
0 = Combined Link/ACT
Note:
Registers 0x1B and 0x1C are reserved.
Table 26. Expanded Memory Data Register (Exp_mem_data) address 0x1D (29’d)
Bit
Bit Name
Default
Description
15:0
Expanded Memory
Data
0, RW
Expanded Memory Data: Data to be written to or read from ex-
panded memory. Note that in 8-bit mode, the data resides at the
LSB octet of this register.
Table 27. Expanded Memory Address Register (Exp_mem_addr) address 0x1E (30’d)
Bit
Bit Name
Default
Description
15:0
Expanded Memory
Address
0, RW
Expanded Memory Address: Pointer to the address in expand-
ed memory. The pointer is 16-bit wide.
www.national.com
38
2.0 Register Block (Continued)
Table 28. PHY Support Register #2 (PHY_SUP) address 0x1F (31’d)
Bit
15:5
4:0
Bit Name
Reserved
Default
Description
Write as 0, ignore on read.
0, RO
PHY Address
STRAP[0_0001], PHY Address: Defines the port on which the PHY will accept Se-
RW rial Management accesses.
39
www.national.com
3.0 Configuration
This section includes information on the various configura-
tion options available with the DP83865. The configuration
options include:
There are three registers used for accessing the expanded
memory. The Expanded Memory Access Control resiger
(0x16) sets up the memory access mode, for example, 8-
bit or 16-bit data addess, enable or disable automatic
address increment after each access, and read/write or
write-only opeation. The Expanded Memory Address
pointer register (0x1E) pionts the location of the expanded
memory to be accessed. The Expanded Memory Data
(0x1D) register contains the data read from or write to the
expanded memory.
— Accessing expanded memory space
— Manual configuration
– Speed / Duplex selection
– Forced Master / Slave
— Auto-Negotiation
– Speed / Duplex selection
– Gigabit speed fallback
– Gigabit retry forced link
– Master / Slave resolution
– Next Page support
Note that the order of the writes to these registers is impor-
tant. While register 0x1E points to the internal expanded
address and register 0x1D contains the data to be written
to or read from the expanded memory, the contents of reg-
ister 0x1E automatically increments after each read or
write to data register 0x1D when auto-increment is
selected. Therefore, if data write need to be confirmed,
address register 0X1E should be reloaded with the original
address before reading from data register 0X1D (when
auto-increment is selected).
– Parallel Detection
– Pause and Asymmetrical Pause resolution
– Restart Auto-Negotiation
– Auto-Negotiation complete time
— Auto-Negotiation register set
— Auto-MDIX configuration
— Automatic polarity correction
— PHY address and LEDs
— Reduced LED mode
The expanded memory space data is 8-bit wide. In the 8-bit
read/write mode, the LSB 8 bits of the data register
0x1D.7:0 is mapped to the expanded memory.
The following is an example of step-by-step precedure
enabling the Speed Fallback mode:
— 1) Power down the DP83865 by setting register 0x00.11
= 1. This is to ensure that the memory access does not
interfere with the normal operation.
— Modulate LED on error
— MII / GMII / RGMII MAC interfaces
— Clock to MAC output
— 2) Write to register 0x16 the value 0x0000. This allows
access to expanded memory for 8-bit read/write.
— MII / GMII /RGMII isolate mode
— Loopback mode
— 3) Write to register 0x1E the value 0x1C0.
— 4) Write to register 0x1D the value 0x0008.
— IEEE 802.3ab test modes
— Interrupt
— 5) Take the out of power down mode by resetting register
0x00.11.
— Power down modes
3.2 Manual Configuration
— Low power mode
For manual configuration of the speed and the duplex
modes (also referred to as forced mode) , the Auto-Negoti-
ation function has to be disabled. This can be done in two
ways. Strapping Auto-Negotiation Enable (AN_EN) pin low
disables the Auto-Negotiation. Auto-Negotiation can also
be disabled by writing a “0” to bit 12 of the BMCR 0x00 to
override the strapping option.
— BIST usage
— Cable length indicator
— 10BASE-T HDX loopback disable
— I/O Voltage Selection
— Non-compliant interoperability mode
The DP83865 supports six different Ethernet protocols:
10BASE-T Full Duplex and Half Duplex, 100BASE-TX
Full Duplex and Half Duplex, 1000BASE-T Full Duplex and
Half Duplex. There are three ways to select the speed and
duplex modes, i.e. manual configuration with external
strapping options or through management register write
and Auto-Negotiation.
It should be noted that manual 1000BASE-T mode is not
supported by IEEE. The DP83865, when in manual
1000BASE-T mode, only communicates with another
National PHY. The manual 1000BASE-T mode is designed
for test purposes only.
3.2.1 Speed/Duplex Selection
In Manual mode, the strapping value of the SPEED[1:0]
pins is used to determine the speed, and the strap value of
the DUPLEX pin is used to determine duplex mode.
3.1 Accessing Expanded Memory Space
The 32 IEEE base page registers limits the number of func-
tions and features to be accessed. The advanced propri-
etary features are implemented in the register located in
the expanded memory space. The following are features
and functions require access to expanded memory space:
For all of the modes above, the DUPLEX strap value “1”
selects Full Duplex (FD), while “0” selects Half Duplex
(HD). The strap values are latched on during power-on
reset and can be overwritten by access to the BMCR regis-
ter 0x00 bits 13,12, 8 and 6.
— Gigabit Speed Fallback
— Gigabit Retry Forced Link
— Cable length indicator
— 10BASE-T HDX loopback
www.national.com
40
3.0 Configuration (Continued)
.
Table 29. Speed/Duplex Selection, AN_EN = 0
Table 31. Master/Slave Resolution, AN_EN = 0
DP83865
Advertise
Link Partner
Advertise
DP83865
Outcome
Link Partner
Outcome
DUPLEX SPEED[1] SPEED[0]
Manual Mode
10BASE-T HD
0
0
0
0
0
1
0
1
0
Manual
Master
Manual
Master
Unresolved
No Link
Unresolved
No Link
100BASE-TX HD
Manual
Master
Manual
Slave
Master
Master
Master
Slave
Slave
Slave
Slave
Master
1000BASE-T HD
(Between National
PHYs only)
Manual
Master
Multi-node
(Auto-neg)
Manual
Master
Single-node
(Auto-neg)
0
1
1
1
1
0
0
1
1
0
1
0
Reserved
10BASE-T FD
100BASE-TX FD
Manual
Slave
Manual
Master
Manual
Slave
Manual
Slave
Unresolved
No Link
Unresolved
No Link
1000BASE-T FD
(Between National
PHYs only)
Manual
Slave
Multi-node
(Auto-neg)
Slave
Master
Manual
Slave
Single-node
(Auto-neg)
Slave
Master
1
1
1
Reserved
3.2.2 Master/Slave
The DP83865 also supports features such as:
— Next Page
In 1000BASE-T the two link partner devices have to be
configured, one as Master and the other as Slave. The
Master device by definition uses a local clock to transmit
data on the wire; the Slave device uses the clock recov-
ered of the incoming data from the link partner for transmit-
ting its data. The Master and Slave assignments can be
manually set by using strapping options or register writes.
When the AN_EN pin is strapped low, strapping MULTI_EN
pin low selects Slave and high selects Master mode. Reg-
ister 9 bits 12:11 allows software to overwrite the strapping
Master/Slave setting (Table 30). Note that if both the link
partner and the local device are manually given the same
Master/Slave assignment, an error will occur as indicated
in 1KSTSR 0x0A bit 15.
— Parallel Detection for 10/100 Mbps
— Restart Auto-Negotiation through software
3.3.1 Speed/Duplex Selection - Priority Resolution
The Auto-Negotiation function provides a mechanism for
exchanging configuration information between the two
ends of a link segment. This mechanism is implemented by
exchanging Fast Link Pulses (FLP). FLP are burst pulses
that provide the signalling used to communicate the abili-
ties between two devices at each end of a link segment.
For further details regarding Auto-Negotiation, refer to
Clause 28 of the IEEE 802.3u specification. The DP83865
supports six different Ethernet protocols: 10BASE-T Full
Duplex, 10BASE-T Half Duplex, 100BASE-TX Full Duplex,
100BASE-TX Half Duplex, 1000BASE-T Full Duplex, and
1000BASE-T Half Duplex. The process of Auto-Negotiation
ensures that the highest performance protocol is selected
(i.e., priority resolution) based on the advertised abilities of
the Link Partner and the local device. (Table 33)
Table 30. 1000BASE-T Master/Slave Sel., AN_EN = 0
MULTI_EN
Manual Mode
0
1
Slave mode
Master mode
.
Depending on what the link partner is configured to, the
manual Master/Slave mode can be resolved to eight possi-
ble outcomes. Only two National PHYs will be able to link
to each other in manual configuration mode. (Table 32)
Table 32. Master/Slave Resolution, AN_EN = 0
DP83865
Advertise
Link Partner
Advertise
DP83865
Outcome
Link Partner
Outcome
3.3 Auto-Negotiation
Manual
Master
Manual
Master
Unresolved
No Link
Unresolved
No Link
All 1000BASE-T PHYs are required to support Auto-Nego-
tiation. (The 10/100 Mbps Ethernet PHYs had an option to
support Auto-Negotiation, as well as parallel detecting
when a link partner did not support Auto-Neg.) The Auto-
Negotiation function in 1000BASE-T has three primary pur-
poses:
Manual
Master
Manual
Slave
Master
Master
Master
Slave
Slave
Slave
Slave
Master
Manual
Master
Multi-node
(Auto-neg)
Manual
Master
Single-node
(Auto-neg)
— Auto-Negotiation of Speed & Duplex Selection
— Auto-Negotiation of Master/Slave Resolution
Manual
Slave
Manual
Master
Manual
Slave
Manual
Slave
Unresolved
No Link
Unresolved
No Link
— Auto-Negotiation of Pause/Asymetrical Pause Resolu-
tion
Manual
Slave
Multi-node
(Auto-neg)
Slave
Master
Manual
Slave
Single-node
(Auto-neg)
Slave
Master
41
www.national.com
3.0 Configuration (Continued)
The default for AN Speed Fallback is that after five tries to
achieve a stable link, the link speed will drop down to the
next lower advertised speed. The default CRC and IE
Speed Fallback is that after five link drops due to increase
error rate, the link speed drops down to the next lower
advertised speed. If during the link retry stage that the link
partner drops the link or the CAT5 cable is unplugged, the
retry counter will reload the retry count with the default
value of five.
Table 33. Speed/Duplex Selection, AN_EN = 1
DUP Speed[1] Speed[0]
Comments
1000/100/10 HDX
1000/100 HDX
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
1000 HDX
Note that the Speed Fallback works only from gigabit mode
to 100 Mbps or 10 Mbps.
1000/10 HDX
1000/100/10 FDX + HDX
1000/100 FDX + HDX
1000 FDX + HDX
1000/10 FDX + HDX
3.3.3 Gigabit Retry Forced Link
Under the situations that the cable media may not be
appropriate for the gigabit transmission, it may take exces-
sive number of retries to achieve a stable link. If achieving
a stable link is the highest priority, the Retry Forced Link
Mode can be enabled. Retry Forced Link Mode allows
auto-negotiation to force link at the highest common link
speed after five retries.
The Auto-Negotiation priority resolution are as follows:
1. 1000BASE-T Full Duplex (Highest Priority)
2. 1000BASE-T Half Duplex
There are two criteria established to initiate the gigabit
Retry Forced Link.
3. 100BASE-TX Full Duplex
4. 100BASE-TX Half Duplex
1. CRC error rate
2. Idle error rate
5. 10BASE-T Full Duplex
6. 10BASE-T Half Duplex (Lowest Priority)
There are three basic control register bits used to configure
the Speed Fallback and Retry Forced Link. Expanded reg-
ister 0x1C0.3 = 0 enables the Retry Forced Link mode (i.e.,
teh default mode upon power up). LED Control Register
0x13.1:0 selects the criteria for the Speed Fallback. Since
Retry Forced Link does not work when AN fails to achieve
stable link, LED Control Register 0x13.2 should be 0.
3.3.2 Gigabit Speed Fallback
When gigabit mode is advertised, the default auto-negotia-
tion mode attempts to establish link at the highest common
denominator advertised. However, there are situations that
the cable media may not be appropriate for the gigabit
speed communication. If achieving a quality link is the
highest priority, the Speed Fallback Mode can be enabled.
The Speed Fallback Mode allows auto-negotiation to link at
the next lower speed advertised (100Mbps or 10Mbps) if
the gigabit mode fails.
Table 35. LED Control Reg 0x13, Reg 0x1C0.3 = 0
Bit 2, AN Bit 1, CRC Bit 0, IE
Comments
0
0
0
No Retry Forced Link
(RFL)
There are three criteria established to initiate the gigabit
Speed Fallback.
0
0
0
0
1
1
1
0
1
RFL on idle error
1. Auto-negotiation failing to achieve a stable gigabit link
2. CRC error rate
RFL on CRC error
RFL on idle and CRC
3. Idle error rate
There are four basic control register bits used to configure
the Speed Fallback. Expanded register 0x1C0.3 = 1
enables the Speed Fallback mode. LED Control Register
0x13.2:0 selects the criteria for the Speed Fallback.
The default CRC and IE Retry Forced Link is that after five
link drops due to increase error rate, the link will be forced
at the highest advertised speed. If during the link retry
stage that the link partner drops the link or the CAT5 cable
is unplugged, the retry counter will reload the retry count
with the default value of five. Note that the retry may take
forever to achieve a forced link when link partner drops the
link or CAT5 cable is unplugged.
Table 34. LED Control Reg 0x13, Reg 0x1C0.3 = 1
Bit 2, AN Bit 1, CRC Bit 0, IE
Comments
No Speed Fallback (SF)
SF on idle error
0
0
0
0
1
1
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
3.3.4 Master/Slave Resolution
If 1000BASE-T mode is selected during the priority resolu-
tion, the second goal of Auto-Negotiation is to resolve Mas-
ter/Slave configuration. The Master mode priority is given
to the device that supports multiport nodes, such as
switches and repeaters. Single node devices such as DTE
or NIC card takes lower Master mode priority.
SF on CRC error
SF on idle and CRC
SF on failing AN
SF on AN and IE
MULTI_EN is a strapping option for advertising the Multi-
node functionality. (Table 36) In the case when both PHYs
advertise the same option, the Master/Slave resolution is
SF on AN and CRC
SF on AN, CRC, and IE
www.national.com
42
3.0 Configuration (Continued)
resolved by a random number generation. See IEEE
802.3ab Clause 40.5.1.2 for more details.
ANNPTR 0x07 allows for the configuration and transmis-
sion of the Next Page. Refer to clause 28 of the IEEE
802.3u standard for detailed information regarding the
Auto-Negotiation Next Page function.
Table 36. 1000BASE-T Single/Multi-Node, AN_EN = 1
MULTI_EN
Forced Mode
3.3.7 Parallel Detection
0
1
Single node, Slave priority mode
Multi-node, Master priority mode
The DP83865 supports the Parallel Detection function as
defined in the IEEE 802.3u specification. Parallel Detection
requires the 10/100 Mbps receivers to monitor the receive
signal and report link status to the Auto-Negotiation func-
tion. Auto-Negotiation uses this information to configure
the correct technology in the event that the Link Partner
does not support Auto-Negotiation, yet is transmitting link
signals that the 10BASE-T or 100BASE-X PMA recognize
as valid link signals.
Depending on what link the partner is configured to, the
Auto-Negotiation of Master/Slave mode can be resolved to
eight possible outcomes. (Table 37)
If the DP83865 completes Auto-Negotiation as a result of
Parallel Detection, without Next Page operation, bits 5 and
7 of ANLPAR 0x05 will be set to reflect the mode of opera-
tion present in the Link Partner. Note that bits 4:0 of the
ANLPAR will also be set to 00001 based on a successful
parallel detection to indicate a valid 802.3 selector field.
Software may determine that the negotiation is completed
via Parallel Detection by reading ‘0’ in bit 0 of ANER 0x06
after the Auto-Negotiation Complete bit (bit 5, BMSR 0x01)
is set. If the PHY is configured for parallel detect mode and
any condition other than a good link occurs, the parallel
detect fault bit will set (bit 4, ANER 0x06).
Table 37. Master/Slave Resolution, AN_EN = 1
DP83865
Advertise
Link Partner
Advertise
DP83865
Outcome
Link Partner
Outcome
Mult-node
Mult-node
Mult-node
Manual
Master
Slave
Master
Manual
Slave
Master
Slave
Multi-node
M/S resolved
by random seed
M/S resolved
by random seed
Mult-node
Single-node
Master
Slave
Slave
Single-node
Manual
Master
Master
Single-node
Manual
Slave
Master
Slave
3.3.8 Restart Auto-Negotiation
If a link is established by successful Auto-Negotiation and
then lost, the Auto-Negotiation process will resume to
determine the configuration for the link. This function
ensures that a link can be re-established if the cable
becomes disconnected and re-connected. After Auto-
Negotiation is completed, it may be restarted at any time by
writing ‘1’ to bit 9 of the BMCR 0x00.
Single-node
Single-node
Multi-node
Slave
Master
Single-node
M/S resolved
by random seed
M/S resolved
by random seed
3.3.5 Pause and Asymmetrical Pause Resolution
When Full Duplex operation is selected during priority reso-
lution, the Auto-Negotiation also determines the Flow Con-
trol capabilities of the two link partners. Flow control was
originally introduced to force a busy station’s Link Partner
to stop transmitting data in Full Duplex operation. Unlike
Half Duplex mode of operation where a link partner could
be forced to back off by simply generating collisions, the
Full Duplex operation needed a mechanism to slow down
transmission from a link partner in the event that the receiv-
ing station’s buffers are becoming full. A new MAC control
layer was added to handle the generation and reception of
Pause Frames. Each MAC Controller has to advertise
whether it is capable of processing Pause Frames. In addi-
tion, the MAC Controller advertises if Pause frames can be
handled in both directions, i.e. receive and transmit. If the
MAC Controller only generates Pause frames but does not
respond to Pause frames generated by a link partner, it is
called Asymmetrical Pause.
A restart Auto-Negotiation request from any entity, such as
a management agent, will cause DP83865 to halt data
transmission or link pulse activity until the break_link_timer
expires (~1500 ms). Consequently, the Link Partner will go
into link fail mode and the resume Auto-Negotiation. The
DP83865 will resume Auto-Negotiation after the
break_link_timer has expired by transmitting FLP (Fast
Link Pulse) bursts.
3.3.9 Enabling Auto-Negotiation via Software
If the DP83865 is initialized upon power-up with Auto-
Negotiation disabled (forced technology) and the user may
desire to restart Auto-Negotiation, this could be accom-
plished by software access. Bit 12 of BMCR 0x00 should
be cleared and then set for Auto-Negotiation operation to
take place.
The advertisement of Pause and Asymmetrical Pause
capabilities is enabled by writing ‘1’ to bits 10 and 11 of
ANAR 0x04. The link partner’s Pause capabilities is stored
ANLPAR 0x05 bits 10 and 11. The MAC Controller has to
read from ANLPAR to determine which Pause mode to
operate. The PHY layer is not involved in Pause resolution
other than simply advertising and reporting of Pause capa-
bilities.
3.3.10 Auto-Negotiation Complete Time
Parallel detection and Auto-Negotiation take approximately
2-3 seconds to complete. In addition, Auto-Negotiation with
next page exchange takes approximately 2-3 seconds to
complete, depending on the number of next pages
exchanged.
Refer to Clause 28 of the IEEE 802.3u standard for a full
description of the individual timers related to Auto-Negotia-
tion.
3.3.6 Next Page Support
The DP83865 supports the Auto-Negotiation Next Page
protocol as required by IEEE 802.3u clause 28.2.4.1.7. The
43
www.national.com
3.0 Configuration (Continued)
3.4 Auto-Negotiation Register Set
During the next page exchange operation, the station man-
ager can not wait till the end of Auto-Negotiation to read
the ANLPAR because the register is used to store both the
base and next pages. The next page content overwrites the
base page content. The station manager needs to closely
monitor the negotiation status and to perform the following
tasks.
The strapping option settings of Auto-Negotiation, speed,
and duplex capabilities that initialized during power-up or at
reset can be altered any time by writing to the BMCR 0x00,
ANAR 0x04 or, to 1KTCR 0x09.
When Auto-Negotiation is enabled, the DP83865 transmits
the abilities programmed in the ANAR 0x04, and 1KTCR
0x09 via FLP Bursts. The following combinations of
10 Mbps,100 Mbps, 1000 Mbps, Half Duplex, and Full
Duplex modes may be selected.
— ANER 0x06 bit 1 is ‘1’ indicates a page is received. Sta-
tion manage reads the base page information from
ANLPAR0x05 and stores the content in the memory.
— After reading the base page information, software needs
to write to ANNPTR 0x07 to load the next page informa-
tion to be sent.
Table 38. Advertised Modes during Auto-Negotiation,
AN_EN = 1
— The operation can be implemented as polled or interrupt
driven. If another page is received by polling bit 1 in the
ANER 0x06 or by interrupt, the station manager reads bit
15 of the ANLPAR indicating the partner has more next
pages to send. If the partner has more pages to send,
ANNPTR needs to be written to load another next page.
SPEED1 SPEED0 DUPLEX
Adertised Modes
1000BASE-T HD, 10BASE-T HD
1000BASE-T HD
1
1
0
0
1
0
1
0
0
0
0
0
1000BASE-T HD, 100BASE-TX HD
1000BASE-T HD, 100BASE-TX HD,
10BASE-T HD
The ANER 0x06 indicates additional Auto-Negotiation sta-
tus. The ANER provides status on:
1
1
0
0
1
0
1
0
1
1
1
1
1000BASE-T FD, 10BASE-T FD
1000BASE-T FD
— A Parallel Detect Fault has occurred (bit 4, ANER 0x06).
— The Link Partner supports the Next Page function (bit 3,
ANER 0x06).
1000BASE-T FD, 100BASE-TX FD
1000BASE-T FD, 100BASE-TX FD,
10BASE-T FD
— The DP83865 supports the Next Page function (bit 2,
ANER 0x06).
— The current page being exchanged by Auto-Negotiation
has been received (bit1, ANER 0x06).
The Auto-Negotiation protocol compares the contents of
the ANLPAR (received from link partner) and ANAR regis-
ters (for 10/100 Mbps operation) and the contents of
1000BASE-T status and control registers, and uses the
results to automatically configure to the highest perfor-
mance protocol (i.e., the highest common denominator)
between the local and the link partner. The results of Auto-
Negotiation may be accessed in registers BMCR 0x00
(Duplex Status and Speed Status), and BMSR 0x01 (Auto-
Neg Complete, Remote Fault, Link).
— The Link Partner supports Auto-Negotiation (bit 0, ANER
0x06).
The ANNPTR 0x07 contains the next page code word to be
transmitted. See also Section “2.3 Register Description”
for details.
3.5 Auto-MDIX resolution
The GigPHYTER V can determine if a “straight” or “cross-
over” cable is used to connect to the link partner. It can
automatically re-assign channel A and B to establish link
with the link partner, (and channel C and D in 1000BASE-T
mode). Auto-MDIX resolution precedes the actual Auto-
Negotiation process that involves exchange of FLPs to
advertise capabilities. Automatic MDI/MDIX is described in
IEEE 802.3ab Clause 40, section 40.8.2. It is not a required
implementation for 10BASE-T and 100BASE-TX.
The BMCR 0x00 provides control for enabling, disabling,
and restarting the Auto-Negotiation process.
The BMSR 0x01 indicates the set of available abilities for
technology types, Auto-Negotiation ability, and extended
register capability. These bits are permanently set to indi-
cate the full functionality of the DP83865. The BMSR also
provides status on:
— Auto-Negotiation is completed on bit 5
— The Link Partner is advertising that a remote fault has
occurred on bit 4
Table 39. PMA signal to MDI and MDIX pin-out
— A valid link has been established on bit 2
Contact
MDI
MDIX
The ANAR 0x04 stores the capabilities advertised during
Auto-Negotiation. All available capabilities are transmitted
by default. However, the advertised capability can be sup-
pressed by writing to the ANAR. This is a commonly used
by a management agent to change (i.e., to force) the com-
munication technology.
1
2
3
4
5
6
7
8
MDI_A+
MDI_A-
MDI_B+
MDI_C+
MDI_C-
MDI_B-
MDI_D+
MDI_D-
MDI_B+
MDI_B-
MDI_A+
MDI_D+
MDI_D-
MDI_A-
MDI_C+
MDI_C-
The ANLPAR 0x05 is used to store the received base link
code word as well as all next page code words during the
negotiation that is transmitted from the link partner.
If Next Page is NOT being used, then the ANLPAR will
store the base link code word (link partner's abilities) and
retain this information from the time the page is received,
indicated by a ‘1’ in bit 1 of the ANER 0x06, through the
end of the negotiation and beyond.
www.national.com
44
3.0 Configuration (Continued)
To enable Auto-MDIX, strapping option pin MDIX_EN
should be pulled up or left floating. Auto-MDIX can be dis-
abled by strapping MDIX_EN pin low. When Auto-MDIX is
disabled, the PMA is forced to either MDI (“straight”) or
MDIX (“crossed”) - according to the setting of the
MAN_MDIX strapping option pin (high for MDIX and low for
MDI).
and it is implemented on DP83865DVH. Note that the
reduced LED mode is in addition to the existing five-LED
mode.
There are two reduced LED modes, the 3-in-1 mode and
the 4-in-1 mode. The 3-in-1 mode combines 10/100/100
Mbps links status in one LED, the standard LINK10_LED.
In the 3-in-1 mode, the rest of the four LED’s would still
function in the standard mode. This would allow user to use
one LED to indicate three-speed links, and other LED’s to
indicate 1000M link, TX/RX activity, or duplex.
The two strapping options for the MDI/MDIX configuration
can be overwritten by writing to bits 14 and 15 of register
AUX_CTRL (0x12). Bit 15 disables the Auto-MDIX feature
and bit 14 can change the straight/crossed and MDI/MDIX
setting.
Similar to 3-in-1 mode, the 4-in-1 mode combines an addi-
tional activity into the three-speed link modes. This mode
would further reduce the number of LED’s and still keep the
same number of display types.
Auto-MDIX is independent of Auto-Negotiation. Auto-MDIX
works in both AN mode and manual forced speed mode.
The Auto-MDIX in forced speed mode is added to
DP83865DVH revision and up.
To enable the RLED mode, LED Control Register 0x13.5 =
1, and register 0x1A.0 selects 3-in-1 or 4-in-1 mode.
3.6 Polarity Correction
Table 40. Reduced LED Mode
The GigPHYTER V will automatically detect and correct for
polarity reversal in wiring between the +/- wires for each
pair of the 4 ports.
RLED Ena
3/4-in-1 Sel
LINK10_LED
10M link
0
0
1
1
0
1
0
1
10M link
The current status of the polarity reversals is displayed in
bit 15:12 of register LINK_AN (0x11).
10/100/1000 link and ACT
10/100/1000 link
3.7 PHY Address, Strapping Options and LEDs
The PHY address can be set through external strapping
resistors. If all PHY address pins are left floating, the PHY
address is defaulted to 01h by internal pull up/down resis-
tors.
3.9 Modulate LED on Error
The DP83865DVH uses ACT LED to display activity under
normal operation. The ACT LED is steady on when there is
Tx or Rx activity. The ACT can also display gigabit idle
error and CRC event. To differentiate ACT LED from nor-
mal Tx/Rx activity, the rate of the blink is faster when error
occurs. To enable the idle error modulation, LED Control
Register 0x13.3 = 1 and to enable CRC error modulation,
0x13.4 = 1.
The PHY address of DP83865 port can be configured to
any of the 31 possible PHY addresses (except 00h which
puts the PHY in isolation mode at power-up). However, if
more than one DP83865 is used on a board and if MDIO is
bused in a system, each of the DP83865’s address must
be different.
PHY address strapping pin “0” is shared with the Duplex
LED pin.
Table 41. LED Control Reg 0x13
Bit 4
Bit 3
Activity LED
Normal ACT
Strap option pins can be left floating which will result in the
default for the particular pin to be set. External pull-up or
pull-down resistors (2kΩ recommended) can be used to
change the pre-set value.
0
0
1
1
0
1
0
1
ACT/Idle error
ACT/CRC error
The state of the strapping option pin inputs is latched (into
Strap_reg 0x10) at system power-on or reset. For further
details relating to the latch-in timing requirements of the
strapping option pins, as well as the other hardware config-
uration pins, refer to section “6.2 Reset Timing” on
page 73.
ACT/Idle error/CRC error
3.10 MAC Interface
The DP83865 MAC interface can be configured to one of
the following different modes:
Some strap option pins are shared with LED output pins.
Since the strapping resistor could be a pull-up or a pull-
down, an adaptive mechansim has been implemented to
simplify the required external circuit. In case the LED/strap-
ping pin is strapped high, the LED drive level is active low.
In case the LED/strapping pin is strapped low, the LED
drive level is active high. See section “5.9 LED/Strapping
Option” on page 67 for details of the recommende external
components.
— MII Mode: Supports 10/100 Mbps MACs.
— GMII Mode: Supports 802.3z compliant 1000 Mbps
MACs.
— RGMII Mode: Supports RGMII version 1.3.
Only one mode is used at a time.
The interface is capable of driving 35 pF under worst condi-
tions. Note that these outputs are not designed to drive
multiple loads, connectors, backplanes, or cables. See
section “5.6 Layout Notes on MAC Interface” on page 66
for design and layout details.
3.8 Reduced LED Mode
The DP83865DVH has a standard five-LED set. In some
applications, it is desirable to use fewer LED’s. The
“reduced LED mode” (RLED) is created to accommodate
the need for combining the LED functions into fewer LED’s
45
www.national.com
3.0 Configuration (Continued)
3.10.1 MII/GMII Interface
Note that upon power up, the clock output is available after
GPHY goes through its internal reset and initialization pro-
cess. The clock output can be interrupted when GPHY is
going through software reset.
The link speed is determined by Auto-Negotiation, by
strapping options, or by register writes. Based on the
speed linked, an appropriate MAC interface is enabled.
3.12 MII/GMII/RGMII Isolate Mode
The DP83865 can be placed into MII/GMII/RGMII Isolate
mode by writing to bit 10 of the BMCR 0x00.
Table 42. Auto-Negotiation Disabled
SPEED[1:0] Link Strapped
Controller I/F
3.12.1 10/100 Mbps Isolate Mode
00
01
10
11
10BASE-T
100BASE-TX
1000BASE-T
reserved
MII
MII
In Isolation Mode, the DP83865 does not respond to
packet data present at TXD[3:0], TX_EN, and TX_ER
inputs and presents a high impedance on the TX_CLK,
RX_CLK, RX_DV, RX_ER, RXD[3:0], COL, and CRS out-
puts. The DP83865 will continue to respond to all manage-
ment transactions through MDIO.
GMII/RGMII
---
Table 43. Auto-Negotiation Enabled
While in Isolate mode, all medium access operations are
disabled.
Link Negotiated
10BASE-T
Controller I/F
MII
3.12.2 1000 Mbps Isolate Mode
100BASE-TX
1000BASE-T
MII
During 1000 Mbps operation, the isolate mode will TRI-
STATE the GMII outputs of the GigPHYTER V. The PHY
also enters into the power down mode. All medium access
operations are halted. The only way to communicate to the
PHY is through MDIO management port.
GMII/RGMII
3.10.2 RGMII Interface
The Reduced Gigabit Media Independent Interface
(RGMII) is a proposed standard by HP and 3Com. RGMII
is an alternative data interface to GMII and MII. RGMII
reduces the MAC interface pin count to 12.
3.13 Loopback Mode
The DP83865 includes a Loopback Test mode for easy
board diagnostics. The Loopback mode is selected through
bit 14 (Loopback) of BMCR 0x00. Writing 1 to this bit
enables MII/GMII transmit data to be routed to the MII/GMII
receive outputs. While in Loopback mode the data will not
be transmitted onto the media. This is true for 10Mbps, 100
Mbps, as well 1000 Mbps data.
The RGMII can be enabled either through strapping option
or MDIO register write. The strapping pins are shared with
CRS/RGMII_SEL0 and TX_CLK/RGMII_SEL1 since CRS
and TX_CLK signals are not used in the RGMII mode.
Table 44. RGMII Strapping for HP mode
In 10BASE-T, 100BASE-TX, 1000BASE-T Loopback mode
the data is routed through the PCS and PMA layers into the
PMD sublayer before it is looped back. Therefore, in addi-
tion to serving as a board diagnostic, this mode serves as
quick functional verification of the device.
Signal
CRS/RGMII_SEL0
TX_CLK/RGMII_SEL1
Pin
40
Strap
0
1
60
3.14 IEEE 802.3ab Test Modes
Table 45. RGMII Strapping for 3COM mode
IEEE 802.3ab specification for 1000BASE-T requires that
the PHY layer be able to generate certain well defined test
patterns on TX outputs. Clause 40 section 40.6.1.1.2 “Test
Modes” describes these tests in detail. There are four test
modes as well as the normal operation mode. These
modes can be selected by writing to the 1KTCR 0x09 as
shown.
Signal
CRS/RGMII_SEL0
TX_CLK/RGMII_SEL1
Pin
40
Strap
1
1
60
To enable RGMII through software, Register AUX_CTL
0x12.13:12 should be “10” or “11” binary. Note that
enabling the RGMII interface disables GMII and MII inter-
faces.
Table 46. IEEE Test Mode Select
bit 15
bit 14
bit 13
Test Mode Selected
= Test Mode 4
3.11 Clock to MAC Enable
1
0
0
0
0
0
1
1
0
0
0
1
0
1
0
The DP83865 has a clock output (pin 85) that can be used
as a reference clock for other devices such as MAC or
switch silicon. The Clock to MAC output can be enabled
through strapping pins.
= Test Mode 3
= Test Mode 2
= Test Mode 1
The Clock to MAC Enable Strap (pin 88) enables the clock
output. The output frequency can be selected between 25
MHz or 125 MHz. The frequency selection strapping pin is
combined with COL (pin 39), CLK_MAC_FRQ.
= Normal Operation
www.national.com
46
3.0 Configuration (Continued)
See IEEE 802.3ab section 40.6.1.1.2 “Test modes” for
more information on the nature of the test modes.
BIST. The receive BIST contains a receive error counter
and receive packet counter and the transmit BIST is used
to generate Ethernet packets.
The DP83865 provides a test clock synchronous to the
IEEE test patterns. The test patterns are output on the MDI
pins of the device and the test clock is output on the
TX_TCLK pin. There are also two support signals available
which are intended to improve the viewability of the test
patterns on an oscilloscope. TX_TRIGGER marks the start
of the test pattern and TX_SYNC_CLK provides and addi-
tional clock. Refer to section “1.6 Device Configuration and
LED Interface” on page 8 for pin numbers.
The BIST can be used to verify operations of all three
speed modes. The speed mode can be established
through auto-negotiation or manual forced mode. The BIST
may also be used in combination with the loopback mode
to verify both the transmit and receive operations of the
physical layer device.
Receive BIST
BIST_CNT displays the upper or lower 16-bit of an internal
32-bit counter. Bit 14 of BIST_CFG2 (bist_cnt_sel) selects
which 16-bit portion is shown while bit 15 of BIST_CFG1
(bist_cnt_type) selects whether the receive packet counter
or the receive error counter is active. The active counter
can be cleared by writing a ‘1’ to bit 14 of BIST_CFG1. The
receive BIST counter is disabled by default and can be
enabled through bit 15 of BIST_CFG2.
TX_TCLK, TX_TRIGGER and TX_SYN_CLK must be
enabled through bits 6 and 7 of register AUX_CTRL (0x12)
before they can be used.
3.15 Interrupt
The DP83865 can be configured to generate an interrupt
on pin 3 when changes of internal status occur. The inter-
rupt allows a MAC to act upon the status in the PHY with-
out polling the PHY registers. The interrupt source can be
selected through the interrrupt register set. This register set
consists of:
The receive BIST can be enabled during normal operation
in order to monitor the incoming data stream. The BIST
operation will not affect the PHY’s performance or behav-
ior.
— Interrupt Status Register (INT_STATUS 0x14)
— Interrupt Mask Register (INT_MASK 0x15)
— Interrupt Clear Register (INT_CLEAR 0x17)
Transmit BIST
The transmit BIST allows the generation of packets with
pseudo-random (PSR9) or user defined content (bit 10 of
BIST_CFG1), different packet lengths (bit 13 of
BIST_CFG1) and variable interframe gap (bit 12 of
BIST_CFG1). Bits 7:0 of BIST_CFG1 contain the content
of the packet as defined by the user if that option has been
chosen.
Upon reset, the interrupt is disabled and the interrupt regis-
ters are cleared. Any interrupt source can be enabled in the
INT_MASK register.
The interrupt pin is active low. When the interrupt signal is
asserted it will remain asserted until the corresponding sta-
tus bit is cleared.
The number of packets to be sent are specified through bits
13:11 of BIST_CFG2. Setting the enable bit in bit 11 of
BIST_CFG1 starts the transmittal. After the last packet was
sent this bit is automatically cleared. In case the ‘continu-
ous transmit’ has been selected the enable bit must be
cleared in order to stop the stream of packets.
The interrupt pin is tri-stated when the interrupt is not
enabled or no interrupt has occured.
The status bits are the sources of the interrupt. These bits
are mapped in INT_STATUS. When the interrupt status bit
is “1”, the interrupt signal is asserted if the corresponding
INT_MASK bit is enabled. An interrupt status bit can be
cleared by writing a “1” to the corresponding bit in
INT_CLEAR. The clear bit returns to “0” automatically after
the interrupt status bit is cleared.
Table 47. BIST Configuration 1 Reg (0x19)
Bit
Function
15
Set active counter:
‘1’ = Receive error counter
‘0’ = Receive packet counter
3.16 Low Power Mode / WOL
The GigPHYTER V supports the Wake on LAN (WOL) fea-
ture of a higher layer device. In order to achive the least
possible power consumption the DP83865 must be put in
10BASE-T mode (Half or Full Duplex). In this mode the
device uses a maximum of 146mW of power.
14
13
‘1’ = Clear counter
Packet length:
‘1’ = 1514 bytes
‘0’ = 60 bytes
3.17 Power Down Mode
12
Interframe gap:
‘1’ = 9.6 µs
‘0’ = 0.096 µs
Register BMCR (0x00) bit 11 puts the GigPHYTER V in
Power Down mode. Writing a ‘1’ to this location causes the
DP83865 to deactivate everything but the management
(MDC / MDIO) interface. During this mode the device con-
sumes the least possible power.
11
10
‘1’ = Enable transmit BIST
Packet type:
‘1’ = PSR9
‘0’ = User defined
3.18 BIST Configuration
The BIST (Built-In Self Test) provides a test interface that
allows to evaluate receive performance and to generate
valid transmit packets. Registers 0x18 (BIST_CNT), 0x19
(BIST_CFG1) and 0x1A (BIST_CFG2) contain the controls
to two distinct BIST functions: Receive BIST and transmit
7:0
User defined packet content.
47
www.national.com
3.0 Configuration (Continued)
During transmit BIST operation the transmit path
(TXD[7:0]) of the GMII / MII is disabled. All generated pack-
ets will be sent out to the MDI path unless the loopback
mode is enabled. In that case the generated packets will be
presented at the receive path (RXD[7:0]) of the GMII / MII.
3.20 10BASE-T Half Duplex Loopback
By default, the 10BASE-T half duplex transmitted packets
are looped back to the receive side. This is a legacy imple-
mentation. However, in the latest MAC or switch design,
the 10 Mbps loopback is desired to be turned off. The 10
Mbps HDX loopback can be disabled in the expanded
memory register 0x1C0.1.
Table 48. BIST Configuration 2 Reg (0x1A)
Bit
15
14
Function
Table 50. 10M FDX Loopback Disable, Reg 0x1C0
‘1’ = Enable counter
Counter selection:
Bit 1
10BASE-T HDX Loopback Mode
10BASE-T HDX loopback enabled
10BASE-T HDX loopback disabled
0
1
‘1’ = upper 16-bit
‘0’ = lower 16-bit
13:11
Number of packets to transmit:
3.21 I/O Voltage Selection
‘000’ = continuous transmit
‘001’ = 1 packet
‘010’ = 10 packets
There are two options for the I/O voltage available. All
IO_VDD pins must be connected to the same power sup-
ply. It can either be 2.5V or 3.3V. The VDD_SEL pin must
be connected to ground in order to select 2.5V or to the
3.3V power supply to select 3.3 V. This pin must be con-
nected directly to the respective power supply and must
not use a pull-up/-down resistor.
‘011’ = 100 packets
‘100’ = 1,000 packets
‘101’ = 10,000 packets
‘110’ = 100,000 packets
‘111’ = 10,000,000 packets
Pin which are effected by IO_VDD, i.e. will be driven at a
different voltage level, are all pin on the GMII/MII interface,
management interface, JTAG interface, clock interface,
device configuration and reset pins.
If BIST is operating the 1000BASE-T mode, active
GTX_CLK is required for the operation.
3.19 Cable Length Indicator
3.22 Non-compliant inter-operability mode
The maximum CAT5 cable length specified in IEEE 802.3
is 100 meters. When cable length extended beyond the
IEEE specified range, bit error rate (BER) will increase due
to the degredation of signal-to-noise ratio. The DP83865
has enough margin built-in to work at extended cable
reach.
In this mode the DP83865 allows with other vendor’s first
generation 1000 Mbps PHYs. National’s DP83865 is com-
pliant to IEEE 802.3ab and optionally inter-operable with
non-compliant PHYs.
To enter non-compliant inter-operability mode the user can
use a 2kΩ resistor on NON_IEEE_STRAP (pin 1) or write
‘1’ to bit 9 of register 0x12.
When a 100BASE-TX or 1000BASE-T link is established,
the cable length is determined from adaptation parameters.
In 100BASE-TX mode, one cable length measurement is
available since there is one receive channel. In 1000BASE-
T mode, four cable length measurements are available
since there are four receive channels. Each measurement
is stored in an 8-bit register in the expanded memory
space. User may choose to take the average of four mea-
surement to achieve more accurate result. The number
stored in the cable length registers are in meters, and the
typical accuracy is ±5 meters.
The non-compliant mode is functional in auto-negotiation
configuration. It is not applicable in manual speed configu-
ration.
Table 49. Cable Length Indicator Registers
Regiters
Length_A
Length_B
Length_C
Length_D
Addr
Description
0x019F
0x01A2
0x01A5
0x01A8
Length, 100/1000 Mbps
Length, 1000 Mbps
Length, 1000 Mbps
Length, 1000 Mbps
The error rate may be used in conjuction with the cable
length measurement to determine if the link is within IEEE
specifications. If the measurement shows that the cable
length exceeds 130 meters, either the cable is too long or
the cable quality is not meeting the CAT5 standard.
www.national.com
48
4.0 Functional Description
The DP83865 is a full featured 10/100/1000 Ethernet Phys-
ical layer (PHY) chip. It consists of a digital 10/100/1000
Mb/s core with a common TP interface. It also has a com-
bined versitle MAC interface that is capable of interfacing
with MII and GMII controller interfaces. In this section, the
following topics are covered:
4.1.2 Data and Symbol Sign Scrambler Word Generator
The word generator uses the Scrn[32:0] to generate further
scrambled values. The following signals are generated:
Sxn[3:0], Syn[3:0], and Sgn[3:0].
The 4-bit Sxn[3:0] and Syn[3:0] values are then sent to the
scrambler bit generator. The 4-bit Sgn[3:0] sign values are
provided to the sign scrambler nibble generator.
— 1000BASE-T PCS Transmitter
— 1000BASE-T PMA Transmitter
— 1000BASE-T PMA Receiver
— 1000BASE-T PCS Receiver
— Gigabit MII (GMII)
4.1.3 Scrambler Bit Generator
This sub block uses the Sxn and Syn signals along with the
tx_mode and tx_enable signals to generate the Scn[7:0],
that is further scrambled based on the condition of the
tx_mode and tx_enable signal. The tx_mode signal indi-
cates sending idles (SEND_I), sending zeros (SEND_Z) or
sending idles/data (SEND_N). The tx_mode signal is gen-
erated by the micro controller function. The tx_enable sig-
nal is either asserted to indicate data transmission is
occurring or deasserted when there is no data transmis-
sion. The PCS Data Transmission Enable state machine
generates the tx_enable signal.
— Reduced GMII (RGMII)
— 10BASE-T and 100BASE-TX Transmitter
— 10BASE-T and 100BASE-TX Receiver
— Media Dependent Interface (MII)
The 1000BASE-T transceiver includes PCS (Physical Cod-
ing Sublayer) Transmitter, PMA (Physical Medium Attach-
ment) Transmitter, PMA Receiver and PCS Receiver. The
1000BASE-T functional block diagram is shown in section
“ Block Diagram” on page 2.
The 8-bit Scn[7:0] signals are then passed onto the data
scrambler functional block.
4.1.4 Data Scrambler
4.1 1000BASE-T PCS Transmitter
The Data Scrambler generates scrambled data by accept-
ing the TxDn[7:0] data from the GMII and scrambling it
based on various inputs.
The PCS transmitter comprises several functional blocks
that convert the 8-bit TXDn data from the GMII to PAM-5
symbols passed onto the PMA function. The block diagram
of the PCS transmitter data path in Figure 2 provides an
overview of each of the architecture within the PCS trans-
mitter.
The data scrambler generates the 8-bit Sdn[7:0] value,
which scrambles the TxDn data based primarily on the Scn
values and the accompanying control signals.
All 8-bits of Sdn[7:0] are passed onto the bit-to-quinary
symbol mapping block, while 2-bits, Sdn[7:6], are fed into
the convolutional encoder.
The PCS transmitter consists of eight sub blocks:
— LFSR (Linear Feedback Shift Register)
— Data scrambler and symbol sign scrambler word gener-
ator
4.1.5 Convolutional Encoder
— Scrambler bit generator
— Data scrambler
The encoder uses Sdn[7:6] bits and tx_enable to generate
an additional data bit, which is called Sdn[8].
— Convolutional encoder
The one clock delayed versions csn-1[1:0] are passed to
the data scrambler block. This Sdn[8] bit is then passed to
the bit-to-symbol quinary symbol mapping function.
— Bit-to-symbol quinary symbol mapping
— Sign scrambler nibble generator
— Symbol sign scrambler
4.1.6 Bit-to-Symbol Quinary Symbol Mapping
The requirements for the PCS transmit functionality are
also defined in the IEEE 802.3ab specification section
40.3.1.3 “PCS Transmit function”.
This block implements the IEEE 802.3ab specification
Tables 40-1 and 40-2 Bit-to-Symbol Mapping for even and
odd subsets. It takes the 9-bit Sdn[8:0] data and converts it
to the appropriate quinary symbols as defined by the
tables.
4.1.1 Linear Feedback Shift Register (LFSR)
The side-stream scrambler function uses a LFSR imple-
menting one of two equations based on the mode of opera-
tion, i.e., a master or a slave. For master operation, the
equation is
The output of this block generates the TAn, TBn, TCn, and
TDn symbols that passed onto the symbol sign scrambler.
gM(x) = 1 + x13 + x33
4.1.7 Sign Scrambler Nibble Generator
For slave operation, the equation is
gS(x) = 1 + x20 + x33
Sign Scrambler Nibble Generator performs some further
scrambling of the sign values Sgn[3:0] that are generated
by the data and symbol sign scrambler word generator.
The sign scrambling is dependent on the tx_enable signal.
The 33-bit data output, Scrn[32:0], of this block is then fed
to the data scrambler and symbol sign scrambler word gen-
erator.
The SnAn, SnBn, SnCn, and SnDn outputs are then passed
onto the symbol sign scrambler function.
49
www.national.com
4.0 Functional Description (Continued)
Sign
Scrambled
PAM-5
Symbols
to PMA
TAn
TBn
TCn
TDn
Bit-to
Quinary Symbol
Mapping
Sxn[3:0]
Scn[7:0]
Sdn[8:0]
Data Scrambler
Data
Scrambler
Bit
Generator
An
Bn
Cn
Dn
Scrambler and
Convolutional
Encoder
and Symbol
LSFR
Scrn[32:0]
Syn[3:0]
Sign Scrambler
Symbol
g
M = 1 + x13 + x33
gS = 1 + x20 + x33
Sign
Scrambler
Word Generator
SnAn
SnBn
SnCn
SnDn
Sign
Scrambler
Nibble
Generator
g(x) = x3 x8
Sgn[3:0]
Input Data
Byte from GMII
TxDn[7:0]
Figure 2. PCS TX Functional Block Diagram
4.3 1000BASE-T PMA Receiver
4.1.8 Symbol Sign Scrambler
Symbol Sign Scrambler scrambles the sign of the TAn,
TBn, TCn, and TDn input values from the bit-to-symbol qui-
nary symbol mapping function by either inverting or not
inverting the signs. This is done as follows:
The PMA Receiver (the “Receiver”) consists of several sub
functional blocks that process the four digitized voltage
waveforms representing the received quartet of quinary
PAM-5 symbols. The DSP processing implemented in the
receiver extracts a best estimate of the quartet of quinary
symbols originated by the link partner and delivers them to
the PCS Receiver block for further processing. There are
four separate Receivers, one for each twisted pair.
An = TAn x SnAn
Bn = TBn x SnBn
Cn = TCn x SnCn
Dn = TDn x SnDn
The main processing sub blocks include:
— Adaptive Equalizer
The output of this block, namely An, Bn, Cn, and Dn, are the
sign scrambled PAM-5 symbols. They are then passed
onto the PMA for further processing.
— Echo and Crosstalk Cancellers
— Automatic Gain Control (AGC)
— Baseline Wander (BLW) Correction
— Slicer
4.2 1000BASE-T PMA Transmitter
The PMA transmit block shown in Figure 3 contains the fol-
lowing blocks:
4.3.1 Adaptive Equalizer
— Partial Response Encoder
— DAC and Line Driver
The Adaptive Equalizer compensates for the frequency
attenuation characteristics which results from the signal
distortion of the CAT-5 cable. The cable has higher attenu-
ates at the higher frequencies and this attenuation must be
equalized. The Adaptive Equalizer is a digital filter with tap
coefficients continually adapted to minimize the Mean
Square Error (MSE) value of the slicer's error signal output.
Continuous adaptation of the equalizer coefficients means
that the optimum set of coefficients will always be achieved
for maximum specified length or lower quality of cable.
4.2.1 Partial Response Encoder
Partial Response (PR) coding (or shaping) is used on the
PAM-5 coded signals to spectrally shape the transmitted
PAM-5 signal in order to reduce emissions in the critical
frequency band ranging from 30 MHz to 60 MHz. The PR
Z-transform implemented is
0.75 + 0.25 Z–1
4.3.2 Echo and Crosstalk Cancellers
The PR coding on the PAM-5 signal results in 17-level PAM-
5 or PAM-17 signal that is used to drive a common
10/100/1000 DAC and line driver. (Without the PR coding
each signal can have 5 levels given by ± 1, ± 0.5 and 0 V. If
all combinations of the 5 levels are used for the present and
previous outputs, then there are 17 unique output levels
when PR coding is used.)
The Echo and Crosstalk Cancellers cancel the echo and
crosstalk produced while transmitting and receiving simul-
taneously. Echo is produced when the transmitted signal
interferes with the received signal on the same wire pair.
Crosstalk is caused by the transmitted signal appearing on
each of the other three wire pairs interfering with the
receive signal on the fourth wire pair. An Echo and
Crosstalk Canceller is needed for each of the wire pairs.
Figure 3 shows the PMA Transmitter and the embedded
PR encoder block with its inputs and outputs. Figure 4
shows the effect on the spectrum of PAM-5 after PR shap-
ing.
4.3.3 Automatic Gain Control (AGC)
The Automatic Gain Control acts upon the output of the
Echo and Crosstalk Cancellers to adjust the receiver gain.
Different AGC methods are available within the chip and
the optimum gain is selected based on the operational
state the chip (master, slave, start-up, etc.).
4.2.2 DAC and Line Driver
The PAM-17 information from the PR encoder is supplied
to a common 10/100/1000 DAC and line driver that con-
verts digitally encoded data to differential analog voltages.
www.national.com
50
4.0 Functional Description (Continued)
PARTIAL RESPONSE PULSE SHAPE CODING
5-LEVEL PAM-5 TO 17-LEVEL PAM
PAM-5
Z -1
SIGN
SCRAMBLER
3-bits/sample
0.75
0.25
DAC
CONTROL
17-LEVEL
PAM-5
10
TABLE
LOOKUP
Manchester/
MLT-3/PAM-17
ANALOG
20-bits/sample
100
5-bits/sample
MUX
1000
DAC
0.75 X(k) + 0.25 X(k-1)
2-bit MLT-3
PMA Transmitter Block
Manchester coding
Figure 3. PMA Transmitter Block
Transmit Spectra
PAM-5 with PR (.75+.25T)
PAM-5
1.200
1.000
0.800
0.600
0.400
0.200
0.000
-0.200
-0.400
10.00
100.00
critical region -- (30MHz -- 60MHz)
Frequency (M Hz)
Figure 4. Effect on Spectrum of PR-shaped PAM-5 coding
4.3.4 Baseline Wander (BLW) Correction
actual voltage input and the ideal voltage level represent-
ing the symbol value. The error output is fed back to the
BLW, AGC, Crosstalk Canceller and Echo Canceller sub
blocks to be used in their respective algorithms.
Baseline wander is the slow variation of the DC level of the
incoming signal due to the non-ideal electrical characteris-
tics of the magnetics and the inherent DC component of
the transmitted waveform. The BLW correction circuit uti-
lizes the slicer error signal to estimate and correct for BLW.
4.4 1000BASE-T PCS Receiver
The PCS Receiver consists of several sub functional
blocks that convert the incoming quartet of quinary sym-
bols (PAM-5) data from the PMA Receiver A, B, C, and D to
8-bit receive data (RXD[7:0]), data valid (RX_DV), and
receive error (RX_ER) signals on the GMII. The block dia-
gram of the 1000BASE-T Functional Block in section
“ Block Diagram” on page 2 provides an overview of the
4.3.5 Slicer
The Slicer selects the PAM-5 symbol value (+2,+1,0,-1,-2)
closest to the voltage input value after the signal has been
corrected for line Inter Symbol Interference (ISI), attenua-
tion, echo, crosstalk and BLW.
The slicer produces an error output and symbol value deci-
sion output. The error output is the difference between the
51
www.national.com
4.0 Functional Description (Continued)
1000BASE-T transceiver and shows the functionality of the
PCS receiver.
4.4.5 Receive State Machine
The state machine operation is defined in IEEE 802.3ab
section 40.3.1.4. In summary, it provides the necessary
receive control signals of RX_DV and RX_ER to the GMII.
In specific conditions defined in the IEEE 802.3ab specifi-
cation, it generates RXD[7:0] data.
The major sub functional blocks of the PCS Receiver
include:
— Delay Skew Compensation
— Delay Skew Control
— Forward Error Correction (FEC)
— Descrambler Subsystem
— Receive State Machine
4.4.6 ADC/DAC/Timing Subsystem
The 1000BASE-T receive section consists of 4 channels,
each receiving IEEE 802.3ab compliant PAM-5 coded data
including Partial Response (PR) shaping at 125 MBaud
over a maximum of a 100 m of CAT-5 cable. The 4 pairs of
receive input pins are AC coupled through the magnetics to
the CAT-5 cable. Each receive pin pair is differentially ter-
minated into an external 100W resistor to match the cable
impedance. Each receive channel consists of a high preci-
sion Analog to Digital data converter (ADC) which quan-
tizes the incoming data into a digital word at the rate of 125
Mb/s. The ADC is sampled with a clock of 125 MHz which
has been recovered from the incoming data stream.
— ADC/DAC/Timing Subsystem
The requirements for the PCS receive functionality are
defined in the IEEE 802.3ab specification in section
40.3.1.4 “PCS Receive function”.
4.4.1 Delay Skew Compensation
This is a mechanism used to align the received data from
the four PMA receivers and to determine the correct spa-
cial ordering of the four incoming twisted pairs, i.e., which
twisted pair carries An, which one carries Bn, etc. The de-
skewed and ordered symbols are then presented to the
Forward Error Correction (FEC) Decoder. The differential
time or time delay skew is due to the differences in length
of each of the four pairs of twisted wire in the CAT-5 cable,
manufacturing variation of the insulation of the wire pairs,
and in some cases, differences in insulation materials used
in the wire pairs. Correct symbol order to the FEC is
required, since the receiver does not have prior knowledge
of the order of the incoming twisted pairs within the CAT-5
cable.
The 1000BASE-T transmit section consists of 4 channels,
each transmitting IEEE 802.3ab compliant 17-level PAM-5
data at 125 M symbols/second. The 4 pairs of transmit out-
put pins are AC coupled through the magnetics to the CAT-
5 cable. Each transmit pin pair is differentially terminated
into an external 100W resistor to match the cable imped-
ance. Each transmit channel consists of a Digital to Analog
data converter (DAC) and line driver capable of producing
17 discrete levels corresponding to the PR shaping of a
PAM-5 coded data stream. Each DAC is clocked with the
internal 125 MHz clock in the MASTER mode, and the
recovered receive clock in the SLAVE mode operation.
4.4.2 Delay Skew Control
The DP83865 incorporates a sophisticated Clock Genera-
tion Module (CGM) which supports 10/100/1000 modes of
operation with an external 25 MHz clock reference (±50
ppm). The Clock Generation module internally generates
multiple phases of clocks at various frequencies to support
high precision and low jitter Clock Recovery Modules
(CRM) for robust data recovery, and to support accurate
low jitter transmission of data symbols in the MASTER and
SLAVE mode operations.
This sub block controls the delay skew compensation func-
tion by providing the necessary controls to allow for com-
pensation in two dimensions. The two dimensions are
referring to time and position. The time factor is the delay
skew between the four incoming data streams from the
PMA RX A, B, C, and D. This delay skew originates back at
the input to the ADC/DAC/TIMING subsystem. Since the
receiver initially does not know the ordering of the twisted
pairs, correct ordering must be determined automatically
by the receiver during start-up. Delay skew compensation
and twisted pair ordering is part of the training function per-
formed during start-up mode of operation.
4.5 Gigabit MII (GMII)
The Gigabit Media Independent Interface (GMII) is
intended for use between Ethernet PHYs and Station Man-
agement (STA) entities and is selected by either hardware
or software configuration. The purpose of GMII is to make
various physical media transparent to the MAC layer.
4.4.3 Forward Error Correction (FEC) Decoder
The FEC Decoder decodes the quartet of quinary (PAM-5)
symbols and generates the corresponding Sdn binary
words. The FEC decoder uses a standard 8 state Trellis
code operation. Initially, Sdn[3:0] may not have the proper
bit ordering, however, correct ordering is established by the
reordering algorithm at start-up.
The GMII Interface accepts either GMII or MII data, control
and status signals and routes them either to the
1000BASE-T, 100BASE-TX, or 10BASE-T modules,
respectively.
4.4.4 Descrambler Subsystem
The descrambler block performs the reverse scrambling
function that was implemented in the transmit section. This
sub block works in conjunction with the delay skew control.
It provides the receiver generated Sdn[3:0] bits for compar-
ison in the delay skew control function.
www.national.com
52
4.0 Functional Description (Continued)
The mapping of the MAC interface is illustrated below in
Table 51.
RGMII
Table 51. GMII/RGMII/MII Mapping
TX_CLK
TD0
GMII
RGMII
MII
GPHY
TD1
TD2
RXD[3:0]
RXD[4:7]
RX_DV
RX_ER
RX_CLK
RX[3:0]
RXD[3:0]
FUNCTIONAL
BLOCK
TD3
TXEN_ER
RCK
RX_DV
RX_ER
RXDV_ER
RX_CLK
RD0
RX_CLK
TX_CLK
TXD[3:0]
RD1
RGMII_SEL1
TX[3:0]
RD2
TXD[3:0]
TXD[4:7]
TX_EN
TX_ER
GTX_CLK
COL
RD3
RXDV_ER
TXEN_ER
TCK
TX_EN
TX_ER
Figure 5. RGMII Signals
4.6.1 1000 Mbps Mode Operation
COL
CRS
CRS
RGMII_SEL0
All RGMII signals are positive logic. The 8-bit data is multi-
plexed by taking advantage of both clock edges. The lower
4 bits are latched on the positive clock edge and the upper
4 bits are latched on trailing clock edge. The control signals
are multiplexed into a single clock cycle using the same
technique.
The GMII interface has the following characteristics:
— Supports 10/100/1000 Mb/s operation
— Data and delimiters are synchronous to clock references
— Provides independent 8-bit wide transmit and receive
data paths
To reduce power consumption of RGMII interface,
TXEN_ER and RXDV_ER are encoded in a manner that
minimize transitions during normal network operation. This
is done by following encoding method. Note that the value
of GMII_TX_ER and GMII_TX_EN are valid at the rising
edge of the clock. In RGMII mode, GMII_TX_ER is
resented on TXEN_ER at the falling edge of the TCK clock.
RXDV_ER coding is implemented the same fashion.
— Provides a simple management interface
— Uses signal levels that are compatible with common
CMOS digital ASIC processes and some bipolar pro-
cesses
— Provides for Full Duplex operation
The GMII interface is defined in the IEEE 802.3z document
Clause 35. In each direction of data transfer, there are Data
(an eight-bit bundle), Delimiter, Error, and Clock signals.
GMII signals are defined such that an implementation may
multiplex most GMII signals with the similar PCS service
interface defined in IEEE 802.3u Clause 22.
TXEN_ER <= GMII_TX_ER (XOR) GMII_TX_EN
RXDV_ER <= GMII_RX_ER (XOR) GMII_RX_DV
When receiving a valid frame with no error, “RXDV_ER =
True” is generated as a logic high on the rising edge of
RCK and “RXDV_ER = False” is generated as a logic high
at the falling edge of RCK. When no frame is being
received, “RXDV_ER = False” is generated as a logic low
on the rising edge of RCK and “RXDV_ER = False” is gen-
erated as a logic low on the falling edge of RCK.
Two media status signals are provided. One indicates the
presence of carrier (CRS), and the other indicates the
occurrence of a collision (COL). The GMII uses the MII
management interface composed of two signals (MDC,
MDIO) which provide access to management parameters
and services as specified in IEEE 802.3u Clause 22.
When receiving a valid frame with error, “RXDV_ER =
True” is generated as logic high on the rising edge of
RX_CLK and “RXERR = True” is generated as a logic low
on the falling edge of RCK.
The MII signal names have been retained and the functions
of most signals are the same, but additional valid combina-
tions of signals have been defined for 1000 Mb/s operation.
TXEN_ER is treated in a similar manner. During normal
frame transmission, the signal stays at a logic high for both
edges of TCK and during the period between frames where
no error is indicated, the signal stays low for both edges.
4.6 Reduced GMII (RGMII)
The Reduced Gigabit Media Independent Interface
(RGMII) is designed to reduce the number of pins required
to interconnect the MAC and PHY (Figure 5). To accom-
plish this goal, the data paths and all associated control
signals are reduced and are multiplexed. Both rising and
trailing edges of the clock are used. For Gigabit operation
the clock is 125 MHz, and for 10 and 100 Mbps operation
the clock frequencies are 2.5 MHz and 25 MHz, respec-
tively. Please refer to the RGMII Specification version 1.3
for detailed descriptions.
4.6.2 1000 Mbps Mode Timing
At the time of the publication of RGMII standard version
1.3, there are two different implmentations of RGMII, HP
and 3COM. The difference is in setup and hold timing.
The DP83865 implemented the HP timing. The following is
an explanation of the RGMII interface of the DP83865.
53
www.national.com
4.0 Functional Description (Continued)
1000 Mbps Mode Transmit Path Timing
serial data stream for 100BASE-TX operation. Since the
10BASE-T and 100BASE-TX transmitters are integrated
with the 1000BASE-T, the differential output pins, TD+ /-
are routed to channel A of the AC coupling magnetics.
In the transmit path, the TX signals are the output of the
MAC and input of the PHY. The MAC output has a data to
clock skew of -500 ps to +500 ps in both HP and 3COM
mode. The PHY input, on the receiver side, requires data
to clock input skew between 1.0 ns to 2.6 ns. To meet the
minimum data skew of 1.0 ns at the PHY input while the
MAC output skew is at -500 ps (i.e., the worst case), the
clock signal (RGMII_TCK) needs to be delayed by minu-
mum of 1.5 ns. To meet the maximum data skew of 2.6 ns
at the PHY input while MAC output skew is at +500 ps, the
maximum clock delay (RGMII_TCK) needs to be within 2.1
ns.
The block diagram in Figure 6 provides an overview of
each functional block within the 10BASE-T and 100BASE-
TX transmit section. The Transmitter section consists of the
following functional blocks:
10BASE-T:
— NRZ to Manchester Encoder
— Link Pulse Generator
— Transmit Driver
The 3COM mode clock delay is implemented internal in the
DP83865DVH. The HP or 3COM mode can be selected at
register 0x12.13:12.
— Jabber Detect
100BASE-TX:
— Code-group Encoder and Injection block
— Parallel-to-Serial block
— Scrambler block
1000 Mbps Mode Receive Path Timing
In the data receive path, the RX signals are the output of
the PHY and input of the MAC. The PHY output has a data
to clock skew of -500 ps to +500 ps (i.e., the HP mode).
— NRZ to NRZI encoder block
— Binary to MLT-3 converter / DAC / Line Driver
If the MAC input, on the receiver side, is operating in
3COM mode that requires minimum of 1.0 ns setup time,
the clock signal (RGMII_RX_CLK) needs to be delayed
with minimum of 1.5 ns if the PHY output has a data to
clock skew of -500 ps. The 3COM mode requires the MAC
input has a minimum hold time of 0.8 ns. Meeting the
3COM minimum input hold time, the maximum clock signal
delay while PHY output skew is at +500 ps would be 2.3
ns.
In 10BASE-T mode the transmitter meets the IEEE 802.3
specification Clause 14.
The DP83865 implements the 100BASE-X transmit state
machine diagram as specified in the IEEE 802.3u Stan-
dard, Clause 24.
4.7.1 10BASE-T Manchester Encoder
The encoder begins operation when the Transmit Enable
input (TXE) goes high. The encoder converts the clock and
NRZ data to Manchester data for the transceiver. For the
duration of TXE remaining high, the Transmit Data (TXD) is
encoded for the transmit differential driver. TXD must be
valid on the rising edge of Transmit Clock (TXC). Transmis-
sion ends when TXE goes low. The last transition is always
positive; it occurs at the center of the bit cell if the last bit is
a one, or at the end of the bit cell if the last bit is a zero.
The 3COM mode clock delay is implemented internal in the
DP83865DVH. The HP or 3COM mode can be selected at
register 0x12.13:12.
4.6.3 10/100 Mbps Mode
When RGMII interface is working in the 100 Mbps mode,
the Ethernet Media Independent Interface (MII) is imple-
mented by reducing the clock rate to 25 MHz. For 10 Mbps
operation, the clock is further reduced to 2.5 MHz. In the
RGMII 10/100 mode, the transmit clock RGMII_TX_CLK is
generated by the MAC and the receive clock
RGMII_RX_CLK is generated by the PHY. During the
packet receiving operation, the RGMII_RX_CLK may be
stretched on either the positive or negative pulse to accom-
modate the transition from the free running clock to a data-
synchronous clock domain. When the speed of the PHY
changes, a similar stretching of the positive or negative
pulses is allowed. No glitch is allowed on the clock signals
during clock speed transitions.
4.7.2 Link Pulse Generator
The link generator is a timer circuit that generates a normal
link pulse (NLP) as defined by the 10 Base-T specification
in 10BASE-T mode. The pulse which is 100ns wide is
transmitted on the transmit output, every 16ms, in the
absence of transmit data. The pulse is used to check the
integrity of the connection to the remote MAU.
4.7.3 Transmit Driver
The 10 Mb/s transmit driver in the DP83865 shares the
100/1000 Mb/s common driver.
4.7.4 Jabber Detect
This interface will operate at 10 and 100 Mbps speeds the
same way it does at 1000 Mbps mode with the exception
that the data may be duplicated on the falling edge of the
appropriate clock.
The Jabber Detect function disables the transmitter if it
attempts to transmit a much longer than legal sized packet.
The jabber timer monitors the transmitter and disables the
transmission if the transmitter is active for greater than 20-
30ms. The transmitter is then disabled for the entire time
that the ENDEC module's internal transmit is asserted. The
transmitter signal has to be deasserted for approximately
400-600ms (the unjab time) before the Jabber re-enables
the transmit outputs.
The MAC will hold RGMII_TX_CLK low until it has ensured
that it is operating at the same speed as the PHY.
4.7 10BASE-T and 100BASE-TX Transmitter
Jabber status can be read from BMSR 0x01.1. For 100
Mb/s and 1000 Mb/s operations, Jabber Detect function is
not incorporated so that BMSR 0x01.1 always returns “0”.
The 10BASE-T and 100BASE-TX transmitter consists of
several functional blocks which convert synchronous 4-bit
nibble data, as provided by the MII, to a 10 Mb/s MLT sig-
nal for 10BASE-T operation or scrambled MLT-3 125 Mb/s
www.national.com
54
4.0 Functional Description (Continued)
10BASE-T
TXD[3:0] / TX_ER
TX_CLK
DIVIDER
100BASE-T
TXD[3:0] / TX_ER
4B/5B ENCODER
AND
INJECTION LOGIC
NRZ TO
MANCHESTER
DECODER
FROM PGM
LINK PULSE
GENERATOR
PARALLEL
TO SERIAL
SCRAMBLER
NRZ-TO-NRZI
100BASE-X
LOOPBACK
BINARY-TO-MLT
10, 100, 1000
MUX/DAC/DRIVER
MDI +/−
Figure 6. 10BASE-T/100BASE-TX Transmit Block Diagram
4.7.5 100BASE-T Code-group Encoding and Injection
4.7.6 Parallel-to-Serial Converter
The code-group encoder converts 4-bit (4B) nibble data
generated by the MAC into 5-bit (5B) code-groups for
transmission. This conversion is required to allow control
data to be combined with packet data code-groups. Refer
to Table 52 for 4B to 5B code-group mapping details.
The 5-bit (5B) code-groups are then converted to a serial
data stream at 125 MHz.
4.7.7 Scrambler
The scrambler is required to control the radiated emissions
at the media connector and on the twisted pair cable (for
100BASE-TX applications). By scrambling the data, the
total energy launched onto the cable is randomly distrib-
uted over a wide frequency range. Without the scrambler,
energy levels at the PMD and on the cable could peak
beyond FCC limitations such as frequencies related to
repeating 5B sequences (e.g., continuous transmission of
IDLEs).
The code-group encoder substitutes the first 8-bits of the
MAC preamble with a /J/K/ code-group pair (11000 10001)
upon transmission. The code-group encoder continues to
replace subsequent 4B preamble and data nibbles with
corresponding 5B code-groups. At the end of the transmit
packet, upon the deassertion of Transmit Enable signal
from the MAC, the code-group encoder injects the /T/R/
code-group pair (01101 00111) indicating the end of frame.
After the /T/R/ code-group pair, the code-group encoder
continuously injects IDLEs into the transmit data stream
until the next transmit packet is detected (reassertion of
Transmit Enable).
The scrambler is configured as a closed loop linear feed-
back shift register (LFSR) with an 11-bit polynomial. The
output of the closed loop LFSR is X-ORed with the serial
NRZ data from the serializer block. The result is a scram-
bled data stream with sufficient randomization to decrease
radiated emissions at certain frequencies by as much as 20
55
www.national.com
4.0 Functional Description (Continued)
dB. The DP83865 uses the PHYADDR[4:0] value to set a
unique seed value for the scramblers. The resulting energy
generated by each channel is out of phase with respect to
each channel, thus reducing the overall electro-magnetic
radiation.
Table 52. 4B5B Code-Group Encoding/Decoding
Name PCS
5B
MII 4B Nibble Code
Code-
group
4.7.8 NRZ to NRZI Encoder
DATA CODES
After the transmit data stream has been serialized and
scrambled, the data is NRZI encoded to comply with the
TP-PMD standard for 100BASE-TX transmission over Cat-
egory-5 unshielded twisted pair cable. There is no ability to
bypass this block within the DP83865.
0
1
2
3
4
5
6
7
8
9
A
B
C
D
E
F
11110
01001
10100
10101
01010
01011
01110
01111
10010
10011
10110
10111
11010
11011
11100
11101
0000
0001
0010
0011
0100
0101
0110
0111
1000
1001
1010
1011
1100
1101
1110
1111
IDLE AND CONTROL CODES
H
I
00100 HALT code-group - Error code
11111 Inter-Packet IDLE - 0000 (Note 1)
11000 First Start of Packet - 0101 (Note 1)
10001 Second Start of Packet - 0101 (Note 1)
01101 First End of Packet - 0000 (Note 1)
00111 Second End of Packet - 0000 (Note 1)
J
K
T
R
INVALID CODES
V
V
V
V
V
V
V
V
V
V
00000 0110 or 0101 (Note 2)
00001 0110 or 0101 (Note 2)
00010 0110 or 0101 (Note 2)
00011 0110 or 0101 (Note 2)
00101 0110 or 0101 (Note 2)
00110 0110 or 0101 (Note 2)
01000 0110 or 0101 (Note 2)
01100 0110 or 0101 (Note 2)
10000 0110 or 0101 (Note 2)
11001 0110 or 0101 (Note 2)
Note 1: Control code-groups I, J, K, T and R in data fields will be
mapped as invalid codes, together with RX_ER asserted.
Note 2: Normally, invalid codes (V) are mapped to 6h on RXD[3:0] with
RX_ER asserted.
www.national.com
56
4.0 Functional Description (Continued)
4.7.9 MLT-3 Converter / DAC / Line Driver
phased logic one events. These two binary streams are
then passed to a 10/100/1000 DAC and line driver which
converts the pulses to suitable analog line voltages. Refer
to Figure 8.
The Binary to MLT-3 conversion is accomplished by con-
verting the serial NRZI data stream output from the NRZI
encoder into two binary data streams with alternately
NRZI_in
MLT-3_plus
MLT-3_minus
differential MLT-3
10, 100, 1000
PAM-17_in
20
MLT-3+
Line
Manchester/
MUX DAC
MLT-3
MLT-3-
NRZI_in
Driver
MLT-3/PAM-17
Converter
Manchester
Figure 7. NRZI to MLT-3 conversion
The 100BASE-TX MLT-3 signal sourced by the MDI+/- line
driver output pins is slew rate controlled. This should be
considered when selecting AC coupling magnetics to
ensure TP-PMD Standard compliant transition times (3 ns
< tr < 5 ns).
— Manchester Decoder
— Link Detect
The 100BASE-T Receive section consists of the following
functional blocks:
— ADC Block
The 100BASE-TX transmit TP-PMD function within the
DP83865 outputs only MLT-3 encoded data. Binary data
outputs is not available from the MDI+/- in the 100 Mb/s
mode.
— Signal Detect
— BLW/EQ/AAC Correction
— Clock Recovery Module
— MLT-3 to NRZ Decoder
— Descrambler
4.7.10 TX_ER
Assertion of the TX_ER input while the TX_EN is also
asserted will cause the DP83865 to substitute HALT code-
groups for the 5B data present at TXD[3:0]. However, the
Start-of-Stream Delimiter (SSD) /J/K/ and End-of-Stream
Delimiter (ESD) /T/R/ will not be substituted with HALT
code-groups. As a result, the assertion of TX_ER while
TX_EN is asserted will result in a frame properly encapsu-
lated with the /J/K/ and /T/R/ delimiters which contains
HALT code-groups in place of the data code-groups.
— Serial to Parallel
— 5B/4B Decoder
— Code Group Alignment
— Link Integrity Monitor
Other topics discussed are:
— Bad SSD Detection
— Carrier Sense
— Collision Detect
4.8 10BASE-T and 100BASE-TX Receiver
4.8.1 10BASE-T Receiver
The 10BASE-T receiver converts Manchester codeing to 4-
bit nibble data to the MII. The 100BASE-TX receiver con-
sists of several sub functional blocks which convert the
scrambled MLT-3 125 Mb/s serial data stream to synchro-
nous 4-bit nibble data that is provided to the MII. The
10/100 Mb/s TP-PMD is integrated with the 1000 Mb/s.
The 10/100 differential input data MDI+/- is routed from
channel B of the isolation magnetics.
The receiver includes differential buffer, offset and gain
compensation. The receiver provides the signal condition-
ing to the Clock and Data Recovery block.
4.8.2 Clock and Data Recovery
The Clock and Data Recovery block separates the
Manchester encoded data stream into internal clock sig-
nals and data. Once the input exceeds the squelch require-
ments, Carrier Sense (CRS) is asserted off the first edge
presented to the Manchester decoder.
See Figure 8 for a block diagram of the 10BASE-T AND
100BASE-TX receive function. It provides an overview of
each functional block within the 10/100 receive section.
4.8.3 Manchester Decoder
The 10BASE-T Receive section consists of the following
functional blocks:
Once the Manchester decoder locks onto the incoming
data stream, it converts Manchester data to NRZ data. The
decoder detects the end of a frame when no more mid-bit
transitions are detected. Within one and a half bit times
— Receiver
— Clock and Data Recovery
57
www.national.com
4.0 Functional Description (Continued)
RXD[3:0] /
RX_ER
RXD[3:0] /
RX_ER
10BASE-T
RX_CLK
100BASE-TX
5B/4B DECODER
LOGIC
4-BIT NIBBLE
DEMUX
DIVIDER
SERIAL
TO
PARALLEL
MANCHESTER
TO NRZ
DECODER
DESCRAMB
LER
MLT-3
TO
NRZ
CLOCK &
DATA
RECOVERY
LINK DETECT
SIGNAL DETECT
AAC
BLW
EQ
CLOCK
RECOVERY
CORRECTN
LINK
DETECT
SIGNAL
DETECT
ADC
RECEIVER
100BASE-TX
10BASE-T
MDI +/−
Figure 8. 10BASE-T/100BASE-T Receive Block Diagram
after the last bit, carrier sense is de-asserted. Receive
clock stays active for at least five more bit times after CRS
goes low, to guarantee the receive timings of the controller.
to Digital Converter (ADC) to allow for Digital Signal Pro-
cessing (DSP) to take place on the received signal.
4.8.6 BLW / EQ / AAC Correction
The aligned NRZ data is then parallized and aligned to 4-
bit nibbles that is presented to the MII.
The digital data from the ADC block flows into the DSP
Block (BLW/EQ/AAC Correction) for processing. The DSP
block applies proprietary processing algorithms to the
received signal and are all part of an integrated DSP
receiver. The primary DSP functions applied are:
4.8.4 Link Detector
In 10 BASE-T mode, the link detection circuit checks for
valid NLP pulses transmitted by the remote link partner. If
valid link pulses are not received the link detector will dis-
able the twisted pair transmitter, receiver and collision
detection functions.
— BLW is defined as the change in the average DC con-
tent, over time, of an AC coupled digital transmission
over a given transmission medium. (i.e. copper wire).
BLW results from the interaction between the low fre-
quency components of a transmitted bit stream and the
frequency response of the AC coupling component(s)
within the transmission system. If the low frequency con-
tent of the digital bit stream goes below the low frequen-
cy pole of the AC coupling transformer then the droop
characteristics of the transformer will dominate resulting
4.8.5 100 BASE-TX ADC Block
The DP83865 requires no external attenuation circuitry at
its receive inputs, MDI+/-. It accepts TP-PMD compliant
waveforms directly from a 1:1 transformer. The analog
MLT-3 signal (with noise and system impairments) is
received and converted to the digital domain via an Analog
www.national.com
58
4.0 Functional Description (Continued)
Figure 9. 100BASE-TX BLW Event
in potentially serious BLW. The digital oscilloscope plot
provided in Figure 9 illustrates the severity of the BLW
event that can theoretically be generated during
100BASE-TX packet transmission. This event consists
of approximately 800 mV of DC offset for a period of 120
ms. Left uncompensated, events such as this can cause
packet loss.
35
30
25
20
15
10
5
150m
130m
— In high-speed twisted pair signalling, the frequency con-
tent of the transmitted signal can vary greatly during nor-
mal operation based primarily on the randomness of the
scrambled data stream and is thus susceptible to fre-
quency dependent attenuation (see Figure 10). This
variation in signal attenuation caused by frequency vari-
ations must be compensated to ensure the integrity of
the transmission. In order to ensure quality transmission
when using MLT-3 encoding, the compensation must be
able to adapt to various cable lengths and cable types
depending on the installed environment. The usage of
long cable length requires significant compensation
which will over-compensate for shorter and less attenu-
ating lengths. Conversely, the usage of short or interme-
diate cable length requiring less compensation will cause
serious under-compensation for longer length cables.
Therefore, the compensation or equalization must be
adaptive to ensure proper level of the received signal in-
dependent of the cable length.
100m
50m
0
0m
0
20
40
60
80
100 120
Frequency (MHz)
Figure 10. EIA/TIA Attenuation vs. Frequency for 0, 50,
100, 130 & 150 meters of CAT 5 cable
— Automatic Attenuation Control (AAC) allows the DSP
block to fit the resultant output signal to match the limit
characteristic of its internal decision block to ensure error
free sampling.
100BASE-TX Standard for both voltage thresholds and tim-
ing parameters.
4.8.7 Signal Detect
Note that the reception of fast link pulses per IEEE 802.3u
Auto-Negotiation by the 100BASE-X receiver will not cause
the DP83865 to assert signal detect.
In 100BASE-TX mode, the link is established by detecting
the scrambled idles from the link partner.
In 100BASE-T mode, the signal detect function of the
DP83865 meets the specifications mandated by the ANSI
FDDI TP-PMD Standard as well as the IEEE 802.3
4.8.8 Clock Recovery Module
The Clock Recovery Module generates a phase corrected
clocks for the 100BASE-T receiver.
59
www.national.com
4.0 Functional Description (Continued)
The CRM is implemented using an advanced digital Phase
Locked Loop (PLL) architecture that replaces sensitive
analog circuitry. Using digital PLL circuitry allows the
DP83865 to be manufactured and specified to tighter toler-
ances.
version ceases upon the detection of the /T/R/ code-group
pair denoting the End of Stream Delimiter (ESD) or with the
reception of a minimum of two IDLE code-groups.
4.8.13 100BASE-X Link Integrity Monitor
The 100BASE-X Link monitor ensures that a valid and sta-
ble link is established before enabling both the Transmit
and Receive PCS layer. Signal Detect must be valid for at
least 500 ms to allow the link monitor to enter the “Link Up”
state, and enable transmit and receive functions.
4.8.9 MLT-3 to NRZ Decoder
The DP83865 decodes the MLT-3 information from the
DSP block to binary NRZI form and finally to NRZ data.
4.8.10 Descrambler
4.8.14 Bad SSD Detection
A serial descrambler is used to de-scramble the received
NRZ data. The descrambler has to generate an identical
data scrambling sequence (N) in order to recover the origi-
nal unscrambled data (UD) from the scrambled data (SD)
as represented in the equations:
A Bad Start of Stream Delimiter (Bad SSD) is any transition
from consecutive idle code-groups to non-idle code-groups
which is not prefixed by the code-group pair /J/K/.
If this condition is detected, the DP83865 will assert
RX_ER and present RXD[3:0] = 1110 to the MII for the
cycles that correspond to received 5B code-groups until at
least two IDLE code groups are detected.
SD= (UD N)
UD= (SD N)
Synchronization of the descrambler to the original scram-
bling sequence (N) is achieved based on the knowledge
that the incoming scrambled data stream consists of
scrambled IDLE data. After the descrambler has recog-
nized 12 consecutive IDLE code-groups, where an
unscrambled IDLE code-group in 5B NRZ is equal to five
consecutive ones (11111), it will synchronize to the receive
data stream and generate unscrambled data in the form of
unaligned 5B code-groups.
Once at least two IDLE code groups are detected, RX_ER
and CRS become de-asserted.
4.8.15 Carrier Sense
Carrier Sense (CRS) may be asserted due to receive activ-
ity once valid data is detected via the Smart squelch func-
tion.
For 10/100 Mb/s Half Duplex operation, CRS is asserted
during either packet transmission or reception.
In order to maintain synchronization, the descrambler must
continuously monitor the validity of the unscrambled data
that it generates. To ensure this, a line state monitor and a
hold timer are used to constantly monitor the synchroniza-
tion status. Upon synchronization of the descrambler the
hold timer starts a 722 ms countdown. Upon detection of
sufficient IDLE code-groups (16 idle symbols) within the
722 ms period, the hold timer will reset and begin a new
countdown. This monitoring operation will continue indefi-
nitely given a properly operating network connection with
good signal integrity. If the line state monitor does not rec-
ognize sufficient unscrambled IDLE code-groups within the
722 ms period, the entire descrambler will be forced out of
the current state of synchronization and reset in order to re-
acquire synchronization.
For 10/100 Mb/s Full Duplex operation, CRS is asserted
only due to receive activity.
CRS is deasserted following an end of packet.
4.8.16 Collision Detect and Heartbeat
A collision is detected on the twisted pair cable when the
receive and transmit channels are active simultaneously
while in Half Duplex mode.
Also after each transmission, the 10 Mb/s block will gener-
ate a Heartbeat signal by applying a 1 us pulse on the COL
lines which go into the MAC. This signal is called the Signal
Quality Error (SQE) and it’s function as defined by IEEE
802.3 is to assure the continued functionality of the colli-
sion circuitry.
4.8.11 Serial to Parallel Converter
The 100BASE-X receiver includes a Serial to Parallel con-
verter this operation also provides code-group alignment,
and operates on unaligned serial data from the descram-
bler (or, if the descrambler is bypassed, directly from the
MLT-3 to NRZ decoder) and converts it into 5B code-group
data (5 bits). Code-group alignment occurs after the /J/K/
code-group pair is detected. Once the /J/K/ code-group
pair (11000 10001) is detected, subsequent data is aligned
on a fixed boundary.
4.9 Media Independent Interface (MII)
The DP83865 incorporates the Media Independent Inter-
face (MII) as specified in Clause 22 of the IEEE 802.3u
standard. This interface may be used to connect PHY
devices to a MAC in 10/100 Mb/s mode. This section
describes both the serial MII management interface as well
as the nibble wide MII data interface.
The serial management interface of the MII allows for the
configuration and control of multiple PHY devices, gather-
ing of status, error information, and the determination of the
type and capabilities of the attached PHY(s).
4.8.12 5B/4B Decoder
The code-group decoder functions as a look up table that
translates incoming 5B code-groups into 4B nibbles. The
code-group decoder first detects the /J/K/ code-group pair
preceded by IDLE code-groups and replaces the /J/K/ with
MAC preamble. Specifically, the /J/K/ 10-bit code-group
pair is replaced by the nibble pair (0101 0101). All subse-
quent 5B code-groups are converted to the corresponding
4B nibbles for the duration of the entire packet. This con-
The nibble wide MII data interface consists of a receive bus
and a transmit bus each with control signals to facilitate
data transfer between the PHY and the upper layer (MAC).
This section covers the follwing subjects:
— Serial Management Register Access
— Serial Management Access Protocol
www.national.com
60
4.0 Functional Description (Continued)
— Serial Management Preample Suppression
— PHY Address Sensing
— MII Data Interface
order to initialize the MDIO interface, the station manage-
ment entity sends a sequence of 32 contiguous logic ones
on MDIO to provide the DP83865 with a sequence that can
be used to establish synchronization. This preamble may
be generated either by driving MDIO high for 32 consecu-
tive MDC clock cycles, or by simply allowing the MDIO pull-
up resistor to pull the MDIO pin high during which time 32
MDC clock cycles are provided. In addition 32 MDC clock
cycles should be used to re-synchronize the device if an
invalid start, op code, or turnaround bit is detected.
— MII Isolate Mode
— Status LED’s
4.9.1 Serial Management Register Access
The serial management MII specification defines a set of
thirty-two 16-bit status and control registers that are acces-
sible through the management interface pins MDC and
MDIO for 10/100/1000 Mb/s operation. The DP83865
implements all the required MII registers as well as several
optional registers. These registers are fully described in
section “2.3 Register Description”. Note that by default, the
PHY base address is 01H that is the Port 1 address. If mul-
tiple PHY’s are used, MDC and MDIO for each DP83865
may be connected together to simplify the interface. The
base address for each single PHY should be different.
The DP83865 operation is pending until it receives the pre-
amble sequence before responding to any other transac-
tion. Once the DP83865 serial management port has been
initialized no further preamble sequencing is required until
after power-on, reset, invalid Start, invalid Opcode, or
invalid turnaround bit occurrs.
The Start code is indicated by a <01> pattern. This assures
the MDIO line transitions from the default idle line state.
Turnaround is defined as an idle bit time inserted between
the register address field and the data field. To avoid con-
tention during a read transaction, no device shall actively
drive the MDIO signal during the first bit of Turnaround.
The addressed DP83865 drives the MDIO with a zero for
the second bit of turnaround and follows this with the
required data. Figure 11 shows the timing relationship
between MDC and the MDIO as driven/received by the Sta-
tion (STA) and the DP83865 (PHY) for a typical register
read access.
4.9.2 Serial Management Access Protocol
The serial control interface consists of two pins, Manage-
ment Data Clock (MDC) and Management Data Input/Out-
put (MDIO). MDC has a maximum clock rate of 2.5 MHz
and no minimum rate. The MDIO line is bi-directional and is
capable of addressing up to thirty-two PHY addresses. The
MDIO frame format is shown below in Table 53.
The MDIO pin requires a pull-up resistor (2 kΩ). During
IDLE and Turnaround, the MDIO signal is pulled high. In
Table 53. Typical MDIO Frame Format
MII Management
Serial Protocol
<idle><start><op code><device addr><reg addr><turnaround><data><idle>
Read Operation
Write Operation
<idle><01><10><AAAAA><RRRRR><Z0><xxxx xxxx xxxx xxxx><idle>
<idle><01><01><AAAAA><RRRRR><10><xxxx xxxx xxxx xxxx><idle>
MDC
Z
Z
MDIO
(STA)
Z
Z
MDIO
(PHY)
Z
Z
Z
0 1 1 0 0 1 1 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 1 0 0 0 0 0 0 0 0
Opcode
(Read)
Register Address
(00h = BMCR)
PHY Address
Register Data
Idle
TA
Idle
Start
(PHYAD = 0Ch)
Figure 11. Typical MDC/MDIO Read Operation
For write transactions, the station management entity
writes data to a PHY address thus eliminating the require-
ment for MDIO Turnaround. The Turnaround time is filled
by the management entity by asserting <10>. Figure 12
shows the timing relationship for a typical MII register write
access.
MAC or other management controller) determines that all
PHY’s in the system support Preamble Suppression by
returning a one in this bit, then the station management
entity need not generate preamble for each management
transaction. A minimum of one idle bit between manage-
ment transactions is required as specified in IEEE 802.3u.
After power-up, the DP83865 requires one idle bit prior to
any management access.
4.9.3 Serial Management Preamble Suppression
The DP83865 supports a Preamble Suppression mode as
indicated by a one in bit 6 of the Basic Mode Status Regis-
ter (BMSR 0x01). If the station management entity (i.e.,
61
www.national.com
4.0 Functional Description (Continued)
MDC
Z
Z
MDIO
(STA)
Z
Z
0 1 0 1 0 1 1 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
PHY Address
Register Address
(00h = BMCR)
Opcode
(Write)
Register Data
Idle
Idle
Start
TA
(PHYAD = 0Ch)
Figure 12. Typical MDC/MDIO Write Operation
4.9.7 Status Information
4.9.4 PHY Address Sensing
The DP83865 provides five PHY address pins to set the
PHY address. The information is latched into the
STRAP_REG 0x10.4:0 at device power-up or reset. The
DP83865 supports PHY Address strapping values
1(<00001>) through 31(<11111>). Note that PHY address 0
by default is the broadcast write address and should not be
used as the PHY address.
There are five LED driver pins associated with each port
indicating status information. Status information include
combined link and speed, duplex, and activity.
LINK10_LED: 10 BASE-T link is established by detecting
Normal Link Pulses separated by 16 ms or by packet data
received.
LINK100_LED: 100BASE-TX link is established when the
PHY receives an signal with amplitude compliant with TP-
PMD specifications. This results in an internal generation
of Signal Detect.
4.9.5 MII Data Interface
Clause 22 of the IEEE 802.3u specification defines the
Media Independent Interface. This interface includes a
dedicated receive bus and a dedicated transmit bus. These
two data buses, along with various control and indicate sig-
nals, allow for the simultaneous exchange of data between
the DP83865 and the upper layer agent (MAC).
LINK1000_LED: 1000BASE-T link is established when
Auto-Negotiation has been completed and reliable recep-
tion of signals has been received from a remote PHY.
Link asserts after the internal Signal Detect remains
asserted for a minimum of 500 ms. Link will de-assert
immediately following the de-assertion of the internal Sig-
nal Detect.
The receive interface consists of a nibble wide data bus
RXD[3:0], a receive error signal RX_ER, a receive data
valid flag RX_DV, and a receive clock RX_CLK for syn-
chronous transfer of the data. The receive clock operates
at 25 MHz to support 100 Mb/s and 2.5 MHz for 10 Mb/s
operation.
ACTIVITY_LED: Activity status indicates the PHY is receiv-
ing data, transmitting data or detecting idle error.
DUPLEX_LED: Duplex indicates that the Gig PHYTER is in
Full-Duplex mode of operation when LED is lit.
The transmit interface consists of a nibble wide data bus
TXD[3:0], a transmit error flag TX_ER, a transmit enable
control signal TX_EN, and a transmit clock TX_CLK oper-
ates at 25 MHz for 100 Mb/s and 2.5 MHz for 10 Mb/s.
Additionally, the MII includes the carrier sense signal CRS,
as well as a collision detect signal COL. The CRS signal
asserts to indicate the reception of data from the network
or as a function of transmit data in Half Duplex mode. The
COL signal asserts as an indication of a collision which can
occur during Half Duplex operation when both a transmit
and receive operation occur simultaneously.
4.9.6 MII Isolate Mode
The DP83865 can be forced to electrically isolate its data
paths from the MII or GMII by setting the BMCR 0x00.10 to
“1”. Clearing BMCR 0x00.10 returns PHY back to normal
operation.
In Isolate Mode, the DP83865 does not respond to packet
data present at TXD, TX_EN, and TX_ER inputs and pre-
sents a high impedance on the TX_CLK, RX_CLK, RX_DV,
RX_ER, RXD, COL, and CRS outputs. The DP83865 will
continue to respond to all serial management transactions
over the MDIO/MDC lines.
The IEEE 802.3u neither requires nor assumes any spe-
cific behavior at the MDI while in Isolate mode. For
DP83685, all MDI operations are halted.
www.national.com
62
5.0 Design Guide
The design guide in conjunction with the Reference Design
Schematics/BOM is intended to provide information to
assist in the design and layout of the DP83865 Gigabit
Ethernet Transceiver. The design guide covers the follow-
ing topics:
5.1 Hardware Reset
The active low RESET pin 33 should be held low for a min-
imum of 150 µs to allow hardware reset. For timing details
see Section 6.2. There is no on-chip internal power-on
reset and the DP83865 requires an external reset signal
applied to the RESET pin.
— Hardware Reset
— Clocks
5.2 Clocks
— Power Supply Decoupling
— Sensitive Supply Pins
— PCB Layer Stacking
The CLOCK_IN pin is the 25 MHz clock input to the
DP83865 used by the internal PLL. This input should come
from a 25 MHz clock oscillator or a crystal. (Check
Section 5.13.1 for component requirements.) When using
a crystal, CLOCK_OUT must be connected to the second
terminal of the crystal. For usage with a oscillator the
CLOCK_OUT pin should be left floating.
— Layout Notes on MAC Interface
— Twisted Pair Interface
— RJ-45 Connections
— Unused Pins / Reserved Pins
— LED/Strapping Configuration
— I/O Voltage Considerations
— Power-up Recommendations
— Compoment Selection
The output of the clock signal requires termination consid-
eration. The termination requirement depends on the trace
length of the clock signal. No series or load termination is
required for short traces less than 3 inches. For longer
traces termination resistors are recommended.
VDD = 3.3 V
C2
CLOCK_OUT
CLOCK_OUT
DP83865
DP83865
VDD
25MHz
EN
RT
25MHz
CLOCK_IN
GND
Zo
CLOCK_IN
C1
RT
(Optional)
Crystal option circuit
Oscillator option circuit
Figure 13. Clock Input Circuit
There are a number of ways to terminate clock traces when
an oscillator is used. The commonly used types are series
and parallel termination. Series termination consumes less
power and it is the recommended termination. The value of
the series termination resistor is chosen to match the trace
characteristics impedance. For example, if the clock
source has an output impedance of 20Ω and the clock
trace has the characteristic impedance Zo = 50Ω then Rs =
50 - 20 = 30Ω. The series source termination Rs should be
placed close to the output of the oscillator.
mended by some crystal vendors. Refer to the vendor’s
crystal datasheet for details.
Adequate and proper decoupling is important to the clock
oscillator performance. A multilayer ceramic chip capacitor
should be placed as close to the oscillator’s VDD pin as
possible to supply the additional current during the tran-
sient switching.
EMI is another consideration when designing the clock cir-
cuitry. The EMI field strength is proportional to the current
flow, frequency, and loop area. By applying series termina-
tion, the current flow is less than parallel termination and
the edge speed is slower, making it desirable for EMI con-
siderations. The loop area is defined as the trace length
times the distance to the ground plane, i.e., the current
return path. Keeping the clock trace as short as possible
reduces the loop area that reduces EMI.
The parallel termination consumes more power than series
termination, and yields faster rise and fall times. The value
of the termination is equal to the trace characteristic imped-
ance, RT = Zo. The parallel termination RT should be
placed close to the CLOCK_IN pin to eliminate reflections.
In cases there are multiple PHY deivces reside on the
same board, it may be cost effective to use one oscillator
with a high speed PLL clock distribution driver. Connecting
multiple clock inputs in a daisy chained style should be
avoided, especially when series termination is applied.
It is best to place the oscillator towards the center of the
PCB rather than at the edge. The radiated magnetic field
tends to be stronger when traces are running along the
PCB edge. If the trace has to run along the edge of the
board, make sure the trace to board edge distance is larger
than the trace to ground plane distance. This makes the
field around the trace more easily coupled to the ground
than radiating off the edge. If the clock trace is placed on
the surface layer, placing a parallel ground trace on each
side of the clock trace localizes the EMI and also prevent
crosstalk to adjacent traces. Burying the clock trace in
No termination is necessary if a crystal is used. The crystal
should be placed as close as possible to the CLOCK pins.
The capacitors C1 and C2 are used to adjust the load
capacitance on these pins. (Figure 13.) The total load
capacitance (C1, C2 and crystal) must be within a certain
range for the DP83865 to function properly (see Table 55
for crystal requirements). The parallel resistor RT is recom-
63
www.national.com
5.0 Design Guide (Continued)
between the ground and VDD plane also minimizes EMI
radiation.
introduce inductive coupling leading to ground bounce.
Connect power and ground pins directly to the planes.
Any through-hole clock oscillator component should be
mounted as flat and as close to the PCB as possible.
Excessive leads should be trimmed. Provide a ground pad
equal or larger than the oscillator foot print on the compo-
nent side of the PCB. Tie this ground pad to the ground
plane through multiple vias. This minimizes the distance to
the ground plane and provide better coupling of the electro-
magnetic fields to the board.
The power supply decouping recommendations may be
perceived conservative. However, for the early prototyp-
ing, please follow the guide lines and recommendations to
assure first time success. To lower the manufacturing cost,
the component count may be reduced by the designer after
careful evaluation and extensive tests on EMI and bit-error-
rate (BER) performance.
5.4 Sensitive Supply Pins
5.3 Power Supply Decoupling
The Analog 1V8_AVDD2 and 1V8_AVDD3 supply are sus-
ceptible to noise and requires special filtering to attenuate
high frequencies. A low pass filter for each of the supply pin
is suggested (Figure 15).
The capacitance between power and ground planes can
provide appreciable power supply decoupling for high edge
rate circuits. This "plane capacitor" has very low ESR and
ESL so that the plane capacitance remains effective at the
frequencies so high that chip capacitors become ineffec-
tive. It is strongly recommended that the PC board have
one solid ground plane and at least one split power plane
with 2.5V and 1.8V copper islands. Ideally the PCB should
have solid planes for each of the supply voltages. The
interplane capacitance between the supply and ground
planes may be maximized by reducing the plane spacing.
In addition, filling unused board areas on signal planes with
copper and connecting them to the proper power plane will
also increase the interplane capacitance.
A 1% 9.76 kΩ resistor is needed to connect to the BG_REF
pin. The connections to this resistor needs to be kept as
short as possible (Figure 15).
Avoid placing noisy digital signal traces near these sensi-
tive pins. It is recommended that the above mentioned
components should be placed before other components.
The 1.8V supplies both the digital core and the analog.
The analog power supply is sensitive to noise. To optimize
the analog performance, it is best to locate the voltage reg-
ulator close to the analog supply pins. Avoid placing the
digital core supply and GMAC in the analog return path.
An example of voltage regulator placement is shown in
Figure 16.
The 2.5V and the 1.8V supply pins are paired with their cor-
responding ground pins. Every other paired supply pins
need to be decoupled with Surface Mount Technology
(SMT) capacitors. It’s recommended that SMT capaci-
tance alternates between 0.01 µF and 0.1 µF so that the
resonance frequencies of the capacitors are "dispersed".
The decoupling capacitors should be placed as close to the
supply pin as possible. For optimal results, connect the
decoupling capacitors directly to the supply pins where the
capacitors are placed 0.010 inch to the power pins. For
lowest ESL and best manufacturability, place the plane
connecting via within 0.010 inch to the SMT capacitor pads
(Figure 14).
Ferrite beads could be used to isolate noisy VCC pins and
preventing noise from coupling into sensitive VCC pins.
This bead in conjunction with the bypass capacitors at the
VCC pins form a low pass filter that prevents the high fre-
quency noise from coupling into the quiet VCC. However,
the use of ferrite beads may yield mixed results when the
inductance resonates with the capacitance. To decrease
the likelihood of resonance, a resistor in parallel with the
ferrite bead may be used. The noise characteristics vary
from design to design. Ferrite beads may not be effective
in all cases. The decision is left to the board designer
based on the evaluation of a specific case.
Decoupling capacitor pad
Via to plane
5.5 PCB Layer Stacking
To route traces for the DP83865 PQFP package, a mini-
mum of four PCB layers is necessary. To meet perfor-
mance requirements,
a six layer board design is
recommended. The following is the layer stacking recom-
mendations for four and six-layer boards.
< 10 mil
Four-layer board (typical application: NIC card):
1. Top layer - signal
2. GND
Via
< 10 mil
3. 3.3 Volt power plane
4. Bottom layer - signal, planes for 1.8 Volt and 2.5 Volt
Figure 14. Place via close to pad.
Bulk capacitance supplies current and maintains the volt-
age level at frequencies above the rate that the power sup-
ply can respond to and below frequencies chip capacitors
are effective. To supply lower speed transient current, a
tantalum 10 µF capacitor for each power plane and each
port should also be placed near the DP83865.
Six-layer board:
1. Top layer - signal
2. 2.5 Volt power plane
3. GND
4. 1.8 Volt power plane
5. Power plane for IO_VDD and/or 3.3 Volt
6. Bottom layer - signal
Lowering the power supply plane and ground plane imped-
ance will also reduce the power supply noise. 1 oz. copper
is recommended for the power and ground planes. Avoid
routing power or ground traces to the supply pins that could
Note that signal traces crossing a plane split should be
avoided (Figure 17). Signal crossing a plane split may
cause unpredictable return path currents and would likely
www.national.com
64
5.0 Design Guide (Continued)
VDD = 1.8 V
DP83865
VDD = 1.8 V
CORE_VDD
18
Ω
1V8_AVDD2
0.1 µF
0.01
0.01
µF
0.01 µF
22 µF
Low pass filter for
1V8_AVDD2 only
GND
GND
1V8_AVDD1
VDD = 2.5 V
µF
IO_VDD
GND
0.1 µF
0.01 µF
Typical supply bypassing
(Near pins of the device)
VDD = 2.5 V
GND
2V5_AVDD2
0.01 µF
2V5_AVDD1
9.76 k
Ω
0.01 µF
GND
BG_REF
GND
1%
VDD = 1.8 V
10
Ω
1V8_AVDD3
GND
22 µF
Low pass filter for
1V8_AVDD3 only
Figure 15. Power Supply Filtering
DP83865 and GMAC PCI NIC Card
PHY
MAC
2.5 1.8
Figure 16. 1.8V voltage regualtor placement.
to result in signal quality failure as well as creating EMI
problems.
65
www.national.com
5.0 Design Guide (Continued)
Termination Requirement
The purpose of the series termination is to reduce reflec-
tions and to improve the signal quality. The board designer
should evaluate the reflection and signal integrity to deter-
mine the need for the termination in each design. As a gen-
eral rule, if the trace length is less than 1/6 of the
equivalent length of the rise and fall times, the series termi-
nation is not needed. The following is an example of calcu-
lating the signal trace length.
Do NOT cross plane split
GND or power plane
The rise and fall times of GMII are in the order of 500 ps for
RX_CLK, and GTX_CLK. Propagation Delay = 170 ps/inch
on a FR4 board. Equivalent length of rise time = (1/6) Rise
time (ps) / Delay (ps/inch) = (1/6) *(500/ 170) = 0.5 inch.
Thus, series termination is not needed for traces less than
0.5 inch long.
Figure 17. Signal crossing a plane split
5.6 Layout Notes on MAC Interface
Trace Impedance
The value of the series termination depends on the driver
output impedance and the characteristic impedance of the
PCB trace. Termination value Rs = characteristic imped-
ance Zo - driver output impedance Ro.
All the signal traces of MII and GMII should be impedance
controlled. The trace impedance reference to ground is 50
Ohms. Uncontrolled impedance runs and stubs should be
kept to minimum.
5.7 Twisted Pair Interface
5.6.1 MII, GMII, and RGMII Interfaces
The Twisted Pair Interface consists of four differential
media dependent I/O pairs (MDI_A, MDI_B, MDI_C, and
MDI_D). Each signal is terminated with a 49.9 Ω resistor.
Figure 18 shows a typical connection for channel A. The
circuitry of channels A, B, C, and D are identical. The MDI
signals are directly connect to 1:1 magnetics. To optimize
the performance, National specifies the key parameters for
the magnetics. Please refer to Section 5.13.2.
MII and GMII are single ended signals. The output of these
signals are capable of driving 35 pF under worst condi-
tions. However, these outputs are not designed to drive
multiple loads, connectors, backplanes, or cables.
The following is a layout guide line for the MDI section.
50-Ohm controlled impedance with respect to chassis GND
50-Ohm controlled impedance with respect to VDD or GND
DP83865
PULSE H-5007
RJ-45
1
A+
A-
MX4+
MX4-
TD4+
TD4-
MDI_A+
MDI_A-
2
3
6
4
5
7
8
VDDA = 2.5 V
VDDA = 2.5 V
B+
B-
C+
C-
49.9 Ω
75
75
Ω
Ω
MCT4
MCT1
TCT4
49.9 Ω
D+
D-
0.01 uF
0.01 uF
1000 pF
3 kV
Circuit Ground
Chassis Ground
Figure 18. Twisted Pair/Magnetics interface (Channel A)
— Place the 49.9 Ω 1% termination resistors as close as
possible to the PHY. Place a 0.01 µF decoupling capac-
itor for each channel between 2.5V plane and ground
close to the termination resistor. Place a 0.01 µF decou-
pling capacitor for each port at the transformer center
tab.
— Each MDI pair should be placed as close as possible in
parallel to minimmize EMI and crosstalk. Each member
of a pair should be matched in length to prevent mis-
match in delay that would cause common mode noise.
— Ideally there should be no crossover or via on the signal
paths.
— All the MDI interface traces should have a charateristic
impedance of 50 Ohms to the GND or 2.5V plane. This
is a strict requirement to minimize return loss.
www.national.com
66
5.0 Design Guide (Continued)
5.8 RJ-45 Connections
— The EMI can be further reduced by placing the traces in
the inner layers and making the outer layers chassis
ground.
The magnetics isolates local circuitry from other equipment
that Ethernet connects to. The IEEE isolation test places
stress on the isolated side to test the dielectic strength of
the isolation. The center tap of the isolated winding has a
"Bob Smith" termination through a 75 Ω resistor and 1000
pF cap to chassis ground. The termination capacitor
should have voltage tolerance of 3 kV (Figure 18).
— Generally, it is a good practice not to overlap the circuit
ground plane with the chassis ground that creates cou-
pling. Instead, make chassis ground an isolated island
and make a void between the chassis and circuit ground.
Place two or three 1206 pads across the chassis and cir-
cuit ground void. This will help when experimentally
choosing the appropriate components to pass EMI emis-
sion test.
To pass EMI compliance tests, there are a few helpful rec-
ommendations to follow.
— The RJ-45 is recommended to have metal shielding that
connects to chassis ground to reduce EMI emission.
5.9 LED/Strapping Option
— The isolated side should have the chassis ground "is-
land" placed. The MDI pairs are placed above a contin-
uous chassis ground plane.
When the LED outputs are used to drive LEDs directly, the
active state of each output driver depends on the logic level
sampled by the corresponding strapping input upon power-
up or reset. For example, if a given LED/strapping pin is
resistively pulled low then the corresponding output is con-
figured as an active high LED driver. Conversely, if a given
LED/strapping pin is resistively pulled high then the corre-
sponding output is configured as an active low LED driver.
Figure 20 is an example of a LED/strapping configuration.
— The MDI pairs are suggested to be routed close together
in parallel to reduce EMI emission and common mode
noise (Figure 19).
Care must be taken when the multi-function LED/strapping
pins are desired to be programmable. Depending on the
strap low or high state, two sets of jumpers could be used
(Figure 20). The left side jumper position is connected for
the high strap option, and the right side position is con-
nected for the low strap option.
Did not maintain parallelism
Avoid stubs
TP
Differential signal pair
The value of all external pull-up and pull-down resistor
should be 2 kΩ in order to make absolutely certain that the
correct voltage level is applied to the pin.
TP
GND or power plane
Figure 19. Differential signal pair
VDDIO = 2.5 V
Header for
jumpers
To LED_pin
Hi
Lo
VDDIO = 2.5 V
Programmable
strap high or low
Strap Low
Strap High
Hi
Lo
LED active high
LED active low
Figure 20. LED/strapping option examples.
This method has the advantage of minimizing component
count and board space. However, it is safer to pull the
unused input pins high or low through a current limiting
resistor. This resistor will prevent excessive current drawn
at the input pin in case there is a defect in the input struc-
ture shorting either VCC or GND to the input. Another
advantage of the protection resistor is to reduce the possi-
bility of latch-up. To reduce component count and to save
5.10 Unused Pins and Reserved Pins
Unused CMOS input pins should not be left floating. Float-
ing inputs could have intermediate voltages halfway
between VCC and ground and, as a consequence, turning
on both the NMOS and the PMOS transistors resulting in
high DC current. Floating inputs could also cause oscilla-
tions. Therefore unused inputs should be tied high or low.
In theory CMOS inputs can be directly tied to VCC or GND.
67
www.national.com
5.0 Design Guide (Continued)
board space, the adjacent unused input pins can be
grouped and tied together with a single resistor.
typically 95% of its nominal voltage varies from design to
design.
The number of unused pins and which pins become
unused pins highly depend on the individual application the
DP83865 is used in. Refer to Section 1.0 for each individ-
ual pin that is not used.
There is no specific requirement for power-up sequence for
the DP83865. However, if it is desirable to control the
power up order, it is theoretically advised to power up
CORE_VDD supply first. If there is no such ability all sup-
plies can be powered up at the same time. There is no
known sequence to date that can cause DP83865 in a
latch-up or lock up condition.
Reserved pins must be left floating.
5.11 I/O Voltage Considerations
In any event, the RESET signal should be held low until
after all power supplies have reached their nominal volt-
ages. See Section 6.2 for additional requirements.
The VDD_SEL_STRAP pin selects which I/O voltage
(IO_VDD) is used in an application. The choice is between
2.5V and 3.3V. If the designer was to choose 2.5V an addi-
tional 3.3V supply could be saved. However, the deciscion
should not be soley based on saving components but
rather on the environment the DP83865 operates in.
5.13 Component Selection
5.13.1 Oscillator
IO_VDD supplies the pins for “MAC Interfaces”, “Manage-
ment Interface”, “JTAG Interface”, “Device Configuration
and LED Interface” and “Reset”. All input pins are either
2.5V or 3.3V compatible. All output pins will have a high
level equal to IO_VDD. The designer must make sure that
all connected devices are compatible with the logic ‘1’ state
of the DP83865 (that is either 2.5V or 3.3V).
The requirements of 25 Mhz oscillators and crystals are
listed in Table 54 and Table 55. Some recommended man-
ufacuturers are listed in Table 56.
In the cases where multiple clock sources with the same
frequency are needed, it is recommended to use a clock
distribution circuit in conjuction with a single freqeuncy
generator. These devices may be obtained from vendors
such as Texas Instrument, Pericom, and Integrated Device
Technology.
If 2.5V IOVDD is selected, do not over drive the GPHY
input with 3.3V logic. The over driving may cause exces-
sive EMI noise and reduce GPHY performance. Over driv-
ing may also cause higher power consumption.
Note that the jitter specification was derived from maximum
capacitance load, worst case supply voltage, and wide
temperature range. The actual allowable jitter number may
be significantly higer when driving the DP83865 clock input
under normal operating conditions. Please consult the
respective vendors for specifics.
5.12 Power-up Recommendations
During power-up, the power supply voltages are not avail-
able immediately but ramp up relatively slow compared to
the clock period of the system clock (CLOCK_IN). How
quickly a supply voltage reaches the “power good” level of
Table 54. 25 MHz Oscillator Requirements
Parameter
Frequency
Min
Typ
Max
-
Units
MHz
ppm
ppm
ns
Condition
-
-
25
-
-
Frequency Tolerance
Frequency Stability
Rise/Fall Time
Jitter (short term)
Jitter (long term)
Symmetry
± 50
± 50
6
0 °C to 70 °C
-
-
1 year aging
-
-
20 - 80 %
-
-
25
200
60
10
-
ps
Cycle-to-cycle, driving 10 pF load
Accumulative over 10 µs
-
-
-
ps
40
-
-
%
Logic 0
-
%
IO_VDD = 2.5 or 3.3V nominal
IO_VDD = 2.5 or 3.3V nominal
Logic 1
90
-
%
Table 55. 25 MHz Crystal Requirements
Parameter
Frequency
Min
Typ
Max
-
Units
MHz
ppm
ppm
pF
Condition
-
-
-
25
-
Frequency Tolerance
Frequency Stability
Load Capacitance
± 50
± 50
40
0 °C to 70 °C
1 year aging
-
-
15
-
Total load capacitance including C1
and C2 (see Section 5.2 for dimension-
ing)
www.national.com
68
5.0 Design Guide (Continued)
Manufacturer
Table 56. Recommended Crystal Oscillators
Description
Part Number
Vite Technology
25 MHz 7.5 x 5 mm Oscillator
VCC1-B2B-25M000
www.viteonline.com
Raltron
www.raltron.com
25 MHz 7.5 x 5 mm Oscillator
25MHz Oscillator
C04305L-25.000MHz
NCH089B3-25.0000
ACSHL-25.0000-E-C-C4
SQ2245V-25.0M-30
Pericom
www.saronix.com
Abracon
www.abracon.com
25MHz Oscillator
Pletronics
25MHz Oscillator
www.pletronics.com
Note: Contact Oscillator manufactures for latest information on part numbers and product specifications. All Oscillators
should be thoroughly tested and validated before using them in production.
69
www.national.com
5.0 Design Guide (Continued)
5.13.2 Magnetics
It is important to select the compoment that meets the
requirements. Per IEEE 802.3ab Clause 40.8, the compo-
nent requirements are listed in Table 57. In addition, the
transformer winding should have the configuration shown
in Figure 21. The recommended magnetics has an isola-
tion transformer followed by a common mode choke to
reduce EMI. There is an additional auto-transformer which
is center tapped. To save board space and reduce compo-
nent count, RJ-45 with integrated magnetics may be used.
TCT
TD+
TD-
MCT
MX+
MX-
The following are magnetics meeting the requirements
(Table 57).
Figure 21. Transformer configuration (1 ch)
Table 57. Magnetics Requirements
Parameter
Turn Ratio
Min
-
Typ
Max
Units
-
Condition
± 1%
1:1
-
Insertion Loss
-
-
-
-1.1
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
dB
Vrms
ns
0.1 - 1 MHz
1 - 60 MHz
60 - 100 MHz
100 - 125 MHz
1 - 30 MHz
30 - 40 MHz
40 - 50 MHz
50 - 80 MHz
80 - 100 MHz
1 - 30 MHz
30 - 60 MHz
60 - 100 MHz
1 - 30 MHz
30 - 60 MHz
60 - 100 MHz
HPOT
-
-0.5
-
-
-1.0
-
-
-1.2
Return Loss
-18
-14.4
-13.1
-12
-10
-43
-37
-33
-45
-40
-35
1,500
-
-
-
-
-
-
-
-
-
-
-
Differential to Common
Rejection Ration
-
-
-
-
-
-
Crosstalk
-
-
-
-
-
-
Isolation
Rise Time
-
-
1.6
-
1.8
-
10 - 90 %
Primary Inductance
350
uH
-
Table 58. Recommended Magnetics
Description
Manufacturer
Bel Fuse, Inc.
Part Number
S558-5999-P3
S558-5999-T3
0843-2B1T-33
LF9203
10/100/1000 Mbps Isolation Transformer
10/100/1000 Mbps Isolation Transformer
10/100/1000 Mbps2X1 Integrated Magnetics
10/100/1000 Mbps Isolation Transformer
www.belfuse.com
Delta
www.delta.tw
Halo
10/100/1000 Mbps Isolation Transformer
10/100/1000 Mbps Isolation Transformer
TG1G-S002NZ
000-7093-37R
www.haloelectronics.com
Midcom
www.haloelectronics.com
Pulse Engineering, Inc.
www.pulseeng.com
10/100/1000 Mbps Isolation Transformer
10/100/1000 Mbps Isolation Transformer
H5007
H5008
Note: Contact Magnetics manufactures for latest part numbers and product specifications. All Magnetics should be thor-
oughly tested and validated before using them in production.
www.national.com
70
6.0 Electrical Specifications
Absolute Maximum Ratings
Recommended Operating Condition
Min Typ Max Units
Supply Voltage IO_VDD
-0.4V to 4.2 V
-0.4V to 2.4V
Supply Voltage CORE_VDD,
1V8_AVDD1, 1V8_AVDD2
Supply Voltage IO_VDD
3.135 3.3 3.465
2.375 2.5 2.625
V
V
Supply Voltage IO_VDD
AnalogVoltages 2V5_AVDD1,
2V5_AVDD2
Supply Voltage 2V5_AVDD1,
2V5_AVDD2
-0.4V to 3.6V
Input Voltage (DCIN)
-0.5V to IO_VDD + 0.5V
-0.5V to IO_VDD + 0.5V
Supply Voltage CORE_VDD
AnalogVoltages 1V8_AVDD1,
1V8_AVDD2
1.71 1.8 1.89
V
Output Voltage (DCOUT
)
Storage Temperature
ESD Protection
-65°C to 150°C
6000V
Ambient Temperature (TA)
0
70
°C
CLK_IN Input Freq. Stability
(over temperature)
-50
+50 ppm
Note: Absolute maximum ratings are those values beyond
which the safety of the device cannot be guaranteed. They
are not meant to imply that the device should be operated
at these limits.
CLK_IN Input Jitter pk-pk
CLK_IN Input Duty Cycle
Center Frequency (fc)
100
60
ps
%
40
25
MHz
Thermal Characteristics
Max
Units
Maximum Case Temperature @ 1.0 W
Theta Junction to Case (Tjc) @ 1.0 W
110
17
°C
°C / W
Theta Junction to Ambient (Tja) degrees Celsius/Watt - No Airflow @ 1.0 W
47
°C / W
6.1 DC Electrical Specification
Symbol
Pin Types
Parameter
Conditions
Min
Typ
Max
Units
V
inputs
IH R/GMII
I
Input High Voltage IO_VDD of
3.3V or 2.5 V
1.7
V
I/O
I/O_Z
VIL R/GMII
inputs
I
Input Low Voltage IO_VDD of
3.3V or 2.5 V
GND
0.9
10
V
µA
µA
V
I/O
I/O_Z
IIH R/GMII
IIL R/GMII
I
Input High Current VIN = IO_VDD
Input Low Current VIN = GND
I/O
I/O_Z
I
10
I/O
I/O_Z
VOH R/GMII
outputs
O,
I/O
I/O_Z
Output High
Voltage
IOH = -1.0 mA
IOL = 1.0 mA
2.1
3.6
0.5
VOL R/GMII
outputs
O,
I/O
I/O_Z
Output Low
Voltage
GND
V
IOZ1 R/GMII
IOZ2 R/GMII
I/O _Z
TRI-STATE
Leakage
VOUT = IO_VDD
VOUT = GND
10
µA
µA
V
I/O_Z
TRI-STATE
Leakage
-10
VIH
non-R/GMII
I
Input High
Voltage
2.0
IO_VDD
I/O
I/O_Z
71
www.national.com
6.0 Electrical Specifications (Continued)
Symbol
VIL
Pin Types
Parameter
Input Low
Conditions
Min
Typ
Max
Units
I
GND
0.8
V
I/O
I/O_Z
Voltage
non-R/GMII
VOH
non-R/GMII
O,
I/O
I/O_Z
Output High
Voltage
IO_VDD = 2.5V
IO_VDD = 3.3V
(IO_VDD -
0.5)
IO_VDD
0.4
V
IOH = -4.0 mA for
both
VOL
non-R/GMII
O,
I/O
I/O_Z
Output Low
Voltage
IOL = 4.0 mA
GND
V
R strap
Strap
I
PU/PD internal
resistor value.
20 - 70
kΩ
CIN1
CMOS Input
Capacitance
8
8
pF
pF
COUT1
O, I/O
I/O_Z
CMOS Output
Capacitance
R0 R/GMII
VOD-10
O, I/O_Z
(MDI)
Output impedance VOUT = IO_VDD / 2
35
Ohm
10 M Transmit
VDIFF
2.2
0.950
0.67
2.5
2.8
1.050
0.82
V peak
differential
VOD-100
(MDI)
(MDI)
100 M Transmit
VDIFF
Note 1
1.0
V peak
differential
VOD-1000
1000 M Transmit
VDIFF
0.745
V peak
differential
Note 1: Guaranteed by design.
www.national.com
72
6.0 Electrical Specifications (Continued)
6.2 Reset Timing
VDD 1.8V (core, analog),
2.5V (I/O, analog),
3.3V (I/O if applicable)
T1
CLK_IN
T2
RESET
MDC
32 clocks
T3
T4
Latch-In of Hardware
Configuration Pins
T5
CLK_TO_MAC
Parameter
Description
Notes
Min
Typ Max Units
T1
Reference clock settle time The reference clock must be stable af-
ter the last power supply voltage has
settled and before RESET is deas-
0
µs
serted. (Note 1)
Pins VDD_SEL and CLK_MAC_EN
are latched in during this time.
T2
T3
T4
Hardware RESET Pulse
Width
Power supply voltages and the refer-
ence clock (CLK_IN) have to be sta-
ble.
150
20
µs
ms
ms
Post RESET Stabilization
time prior to MDC preamble management initialization.
for register accesses
MDIO is pulled high for 32-bit serial
External pull configuration Hardware Configuration Pins are de-
latch-in time from the deas- scribed in the Pin Description section.
20
sertion of RESET
Reset includes external hardware and
internal software through registers.
(Note 2)
T5
CLK_TO_MAC Output Sta- If enabled, the CLK_TO_MAC output,
0 + T1
µs
bilization Time
being independent of RESET, power-
down mode and isolation mode, is
available after power-up.
CLK_TO_MAC is a buffered output
CLK_IN. (Note 1)
Note 1: Guaranteed by design. Not tested.
Note 2: It is recommended to use external pull-up and/or pull-down resistors for each of the hardware configuration pins that provide fast RC time con-
stants in order to latch-in the proper value prior to the pin transitioning to an output driver. Unless otherwise noted in the Pin Description section all
external pull-up or pull-down resistors are recommended to be 2kΩ.
73
www.national.com
6.0 Electrical Specifications (Continued)
6.3 Clock Timing
T7
T7
CLK_IN
T6
T8
Parameter
Description
CLK_IN Duty Cycle
Notes
Min
40
Typ
Max
Units
%
T6
T7
60
CLK_IN tR/tF
10% to 90%
1.0 to 2.5
ns
T8
CLK_IN frequency
(25 MHz +/-50 ppm)
24.99875 25.000000 25.001250
MHz
6.4 1000 Mb/s Timing
6.4.1 GMII Transmit Interface Timing
T9
T13
GTX_CLK
T10
T12
T10
TXD[7:0], TX_EN,
TX_ER
T11
T14
MDI
Begin of Frame
Parameter
Description
Notes
Min Typ Max Units
T9
GTX_CLK Duty Cycle
GTX_CLK tR/tF
40
60
1
%
T10
Note 1,4,5
ns
T11
T12
T13
T14
Setup from valid TXD, TX_EN and TXER to ↑ GTX_CLK
Hold from ↑ GTX_CLK to invalid TXD, TX_EN and TXER
GTX_CLK Stability
Note 2,4
Note 3,4
Note 5
2.0
0.0
ns
ns
-100
+100 ppm
152 ns
GMII to MDI latency
Note 1: tr and tf are measured from VIL_AC(MAX) = 0.7V to VIH_AC(MIN) = 1.9V.
Note 2: tsetup is measured from data level of 1.9V to clock level of 0.7V for data = ‘1’; and data level = 0.7V to.clock level 0.7V for data = ‘0’.
Note 3: thold is measured from clock level of 1.9V to data level of 1.9V for data = ‘1’; and clock level = 1.9V to.data level 0.7V for data = ‘0’.
Note 4: GMII Receiver input template measured with “GMII point-to-point test circuit”, see Test Conditions Section
Note 5: Guaranteed by design. Not tested.
www.national.com
74
6.0 Electrical Specifications (Continued)
6.4.2 GMII Receive Timing
T17
T17
T16
RX_CLK
T15
RXD[7:0]
RX_DV
RX_ER
Valid Data
T18
MDI
Begin of Frame
Parameter
T15
Description
Notes
Min Typ Max Units
↑ RX_CLK to RXD, RX_DV and RX_ER delay
RX_CLK Duty Cycle
Note 2, 3, 4 0.5
40
5.5
60
1
ns
%
T16
T17
RX_CLK tR/tF
Note 1, 4, 5
ns
T18
MDI to GMII latency
384
ns
Note 1: tr and tf are measured from VIL_AC(MAX) = 0.7V to VIH_AC(MIN) = 1.9V.
Note 2: tdelay max is measured from clock level of 0.7V to data level of 1.9V for data = ‘1’; and clock level = 0.7V to.data level 0.7V for data = ‘0’.
Note 3: tdelay min is measured from clock level of 1.9V to data level of 1.9V for data = ‘1’; and clock level = 1.9V to.data level 0.7V for data = ‘0’.
Note 4: GMII Receiver input template measured with “GMII point-to-point test circuit”, see Test Conditions Section.
Note 5: Guaranteed by design. Not tested.
75
www.national.com
6.0 Electrical Specifications (Continued)
6.5 RGMII Timing
6.5.1 Transmit and Receive Multiplexing and Timing
TX [3:0]
TXD[3:0]
TX_EN TX_ER
TXD[3:0] TXD[7:4]
TX_ER
TXD[7:4]
TXEN_ER
TX_EN
TCK
TskewT
Tcyc
RCK
RX [3:0]
RXD[3:0] RXD[7:4] RXD[3:0] RXD[7:4]
RXDV_ER
RX_DV
TholdR
RX_ER
RX_ER
RX_DV
TskewR TsetupR
Tcyc
Parameter
Description
Notes
Note 1
Min Typ Max Units
TskewT
TX to Clock skew (at receiver, PHY), HP mode
TX to Clock skew (at receiver, PHY), 3COM mode
RX to Clock skew (at transmitter, PHY), HP mode
1.0
-900
-500
1.4
1.2
7.2
45
2.0
900
500
ns
ps
ps
ns
ns
ns
%
TskewT
TskewR
TsetupR
TholdR
Tcyc
Note 4
Note 4
Note 4
Note 4
Note 2, 4
Note 3
Note 3
Note 4
RX to Clock setup (at transmitter, PHY), 3COM mode
RX to Clock hold (at transmitter, PHY), 3COM mode
Clock Period
8
8.8
55
TDuty_G
TDuty_T
Tr/Tf
Duty Cycle for gigabit
50
50
Duty Cycle for 10/100 BASE-T
Rise/Fall Time (20 -80%)
40
60
%
1.0
ns
Note 1: The PC board design requires clocks to be routed such that an additional trace delay of greater than 1.5 ns is added to the associated clock signal.
Note 2: For 10 Mbps and 100 Mbps, Tcyc will scale to 400ns +-40ns and 40ns +-4ns.
Note 3: Duty cycle may be stretched or shrunk during speed changes or while transitioning to a received packet’s clock domain as long as minimum duty
cycle is not violated and stretching occurs for no more that three Tcyc of the lowest speed transitioned between.
Note 4: Guaranteed by design. Not tested.
www.national.com
76
6.0 Electrical Specifications (Continued)
6.6 100 Mb/s Timing
6.6.1 100 Mb/s MII Transmit Timing
T21
T20
TX_CLK
TXD[3:0], TX_EN,
TX_ER
T19
T22
MDI
Begin of Frame
Parameter
T19
Description
Notes
Min Typ Max Units
TXD[3:0], TX_EN and TX_ER Setup to ↑ TX_CLK
TXD[3:0], TX_EN and TX_ER Hold from ↑ TX_CLK
TX_CLK Duty Cycle
10
-1
ns
ns
%
T20
T21
40
60
T22
MII to MDI latency
136
ns
6.6.2 100 Mb/s MII Receive Timing
T23
RX_CLK
T24
RXD[3:0]
RX_DV
RX_ER
Valid Data
T25
MDI
Begin of Frame
Parameter
T23
Description
Notes
Min Typ Max Units
RX_CLK Duty Cycle
35
10
65
30
%
ns
ns
T24
↑ RX_CLK to RXD[3:0], RX_DV, RX_ER Delay
T25
MDI to MII latency
288
77
www.national.com
6.0 Electrical Specifications (Continued)
6.7 10 Mb/s Timing
6.7.1 10 Mb/s MII Transmit Timing
T28
T27
TX_CLK
TXD[3:0], TX_EN,
TX_ER
T26
T29
MDI
Begin of Frame
Parameter
T26
Description
Notes
Min Typ Max Units
TXD[3:0], TX_EN and TX_ER Setup to ↑ TX_CLK
TXD[3:0], TX_EN and TX_ER Hold from ↑ TX_CLK
TX_CLK Duty Cycle
100
0
ns
ns
%
T27
T28
40
60
T29
MII to MDI latency
125
ns
6.7.2 10 Mb/s MII Receive Timing
T31
RX_CLK
T30
RXD[3:0]
RX_DV
RX_ER
Valid Data
T32
MDI
Begin of Frame
Parameter
Description
Notes
Min Typ Max Units
T30
↑ RX_CLK to RXD[3:0], RX_DV, RX_ER
Delay
100
300
ns
T31
T32
RX_CLK Duty Cycle
MDI to MII latency
35
65
%
1125
ns
www.national.com
78
6.0 Electrical Specifications (Continued)
6.8 Loopback Timing
GTX_CLK
TX_CLK
TX_EN
TXD[7:0]
TXD[3:0]
Valid Data
CRS
T33
RX_CLK
RX_DV
RXD[7:0]
RXD[3:0]
Valid Data
Parameter
Description
Notes
Min
Typ Max Units
T33
TX_EN to RX_DV Loopback
10 Mb/s
2220
380
ns
100 Mb/s
1000 Mb/s
536
Note: During loopback (all modes) both the TD± outputs remain inactive by default.
79
www.national.com
6.0 Electrical Specifications (Continued)
6.9 Serial Management Interface Timing
MDC
T34
T35
MDIO (output)
MDC
T36
T37
MDIO (input)
Valid Data
Parameter
T34
Description
MDC Frequency
Notes
Min
Typ
Max
2.5
Units
MHz
ns
T35
MDC to MDIO (Output) Delay Time
MDIO (Input) to MDC Setup Time
MDIO (Input) to MDC Hold Time
0
300
T36
10
10
ns
T37
ns
www.national.com
80
6.0 Electrical Specifications (Continued)
6.10 Power Consumption
Symbol
Pin Types
Parameter
Conditions
Min
Typ
Max
Units
I1V8_1000
1V8_AVDD, 1V8 Core_VDD = 1.8V,
0.43
A
Core_VDD cur-
rent
1V8_AVDD = 1.8V,
1000 Mbps FDX
I2V5_1000
2V5_AVDD cur-
rent
2V5_AVDD = 2.5V,
1000 Mbps FDX
0.19
0.01
0.01
0.07
A
A
A
A
I2V5_IO_1000
I3V3_IO_1000
I1V8_100
IO_VDD current
IO_VDD current
IO_VDD = 2.5V,
1000 Mbps FDX
IO_VDD = 3.3V,
1000 Mbps FDX
1V8_AVDD, 1V8 Core_VDD = 1.8V,
Core_VDD cur-
rent
1V8_AVDD = 1.8V,
100 Mbps FDX
I2V5_100
2V5_AVDD,
IO_VDD current
IO_VDD = 2.5V,
2V5_AVDD = 2.5V,
0.06
A
100 Mbps FDX
I2V5_IO_100
IO_VDD current
IO_VDD current
IO_VDD = 2.5V,
100 Mbps FDX
0.01
0.01
A
A
I3V3_IO_100
IO_VDD = 3.3V,
100 Mbps FDX
81
www.national.com
7.0 Frequently Asked Questions
the internal 125 MHz clock generated from the CLOCK_IN
clock to transmit data on the wire. The Slave PHY uses the
clock recovered from the link partner’s transmission as the
transmit clock for all four pairs.
7.1 Do I need to access any MDIO register to start
up the PHY?
A: The answer is no. The PHY is a self contained device.
The initial settings of the PHY are configured by the strap-
ping option at the pins. The PHY will start normal operation
based on the strapping options upon power up or reset.
TX_TCLK: The TX_TCLK is an output of the PHY and can
be enabled to come out on pin 6 (during Test Mode 2 and 3
it is automatically enabled). This is a requirement from the
IEEE 802.3ab specification, Clause 40.6.1.2.5.
7.2 I am trying to access the registers through
MDIO and I got invalid data. What should I do?
This is used for 1000 Mbps transmit activity. It has only one
function:
— It is used in “Test Modes 2 & 3” to measure jitter in the
data transmitted on the wire.
A: There are a number of items that you need to check.
— Make sure the MDC frequency is not greater than 2.5
MHz.
Either the reference clock or the clock recovered from
received data is used for transmitting data; depending on
whether the PHY is in MASTER or SLAVE mode.
TX_TCLK represents the actual clock being used to trans-
mit data.
— Check if the MDIO data line has a 2K pull up resistor and
the line is idling high.
— Verify the data timing against the datasheet.
— Be sure the turn around time (TA) is at least 1 bit long.
7.5 What happens to the TX_CLK during 1000
Mbps operation? Similarly what happens to
RXD[4:7] during 10/100 Mbps operation?
7.3 Why can the PHY establish a valid link but can
not transmit or receive data?
A: TX_CLK is not used during the 1000 Mbps operation,
and the RXD[4:7] lines are not used for the 10/100 opera-
tion. These signals are outputs of the Gig PHYTER V. To
simplify the MII/GMII interface, these signals are driven
actively to a zero volt level. This eliminates the need for
pull-down resistors.
A: PHY is a self contained device. The PHY can establish
link by itself without any MAC and management involve-
ment. Here are some suggestions to isolate the problem.
— Use MDIO management access to configure the BIST
registers to transmit packet. If link partner can receive
data, the problem may lie in the MAC interface.
7.6 What happens to the TX_CLK and RX_CLK
during Auto-Negotiation and during idles?
— Check the MAC transmit timing against the PHY
datasheet.
A: During Auto-Negotiation the Gig PHYTER V drives a 25
MHz clock on the TX_CLK and RX_CLK lines. After a valid
link is established and during idle time, these lines are
driven at 2.5 MHz in 10 Mbps, and at 25 MHz in 100 Mbps
mode. In 1000 Mbps mode only RX_CLK is driven at 125
MHz.
— Verify the receive timing of the MAC device to see if it
matches the PHY datasheet.
— If the PHY receives the data correctly, the activity LED
should turn on.
— Start the debugging at the slower 10 Mbps or 100 Mbps
speed.
— Enable the loopback at register 0x00.14. Verify that you
can receive the data that you transmit.
7.7 Why doesn’t the Gig PHYTER V complete
Auto-Negotiation if the link partner is a forced
1000 Mbps PHY?
7.4 What is the difference between TX_CLK,
TX_TCLK, and GTX_CLK?
A: IEEE specifications define “parallel detection” for 10/100
Mbps operation only. Parallel detection is the name given
to the Auto-Negotiation process where one of the link part-
ners is Auto-Negotiating while the other is in forced 10 or
100 Mbps mode. In this case, it is expected that the Auto-
Negotiating PHY establishes half-duplex link at the forced
speed of the link partner.
A: All the 3 clocks above are related to transmitting data.
However, their functions are different:
TX_CLK: The TX_CLK is an output of the PHY and is part
of the MII interface as described in IEEE 802.3u specifica-
tion, Clause 28.
However, for 1000 Mbps operation this parallel detection
mechanism is not defined. Instead, any 1000BASE-T PHY
can establish 1000 Mbps operation with a link partner in the
following two cases:
This is used for 10/100 Mbps transmit activity. It has two
separate functions:
— It is used to synchronize the data sent by the MAC and
to latch this data into the PHY.
— When both PHYs are Auto-Negotiating,
— It is used to clock transmit data on the twisted pair.
— When both PHYs are forced 1000 Mbps. Note that one
of the PHYs is manually configured as MASTER and the
other is manually configured as SLAVE.
GTX_CLK: The GTX_CLK is an output of the MAC and is
part of the GMII interface as described in IEEE 802.3z
specification, Clause 35.
This is used for 1000 Mbps transmit activity. It has only one
function:
7.8 What determines Master/Slave mode when
Auto-Negotiation is disabled in 1000Base-T
mode?
— It is used to synchronize the data sent by the MAC and
to latch this data into the PHY.
A: Disabling 1000 Base-T Auto-Negotiation forces the PHY
to operate in Master or Slave mode. The selection is
through MULTI_EN pin. Since there is no way of knowing
The GTX_CLK is NOT used to transmit data on the twisted
pair wire. For 1000 Mbps operation, the Master PHY uses
www.national.com
82
7.0 Frequently Asked Questions (Continued)
in advance what mode the link partner is operating, there
could be conflict if both PHY are operating in Master or
both in Slave mode. It is recommended that under normal
operation, AN_EN is enabled.
TC = TJ - Pd(Οc)
Where:
TJ = Junction temperature of the die in oC
TC = Case temperature of the package in oC
Pd = Power dissipated in the die in Watts
Oc = 17oC/watt
7.9 How long does Auto-Negotiation take?
A: Two PHY’s typically complete Auto-Negotiation and
establish 1000 Mbps operation in less than 5 seconds.
1000BASE-T Auto-Negotiation process takes longer than
the 10/100 Mbps. The gigabit negotiation does Next Page
exchanges and extensive line adaptation.
For reliability purposes the maximum junction should be
kept below 120 oC. If the Ambient temperature is 70 oC
and the power dissipation is 1.2 watts then the Maximum
Case Temperature should be maintained at:
TC max = 120oC - 1.1 watts * (17oC/watt)
TC max = 101oC
7.10 How do I measure FLP’s?
A: In order measure FLP’s Auto MDIX function must be
disabled. When in Auto MDIX mode the DP83865 outputs
link pulses every 150 µs. Note that MDIX pulse should not
be confused with the FLP pulses which occur every 125 µs
+/- 14 µs. To disable Auto MDIX, AUX_CTL 0x12.15 = 0.
7.15 The DP83865 will establish Link in 100 Mbps
mode with a Broadcom part, but it will not estab-
lish link in 1000 Mbps mode. When this happens
the DP83865’s Link LED will blink on and off.
Once Auto MDIX is disabled register bit 0x12.14 specifies
MDIX mode. ‘1’ for MDIX cross over mode and ‘0’ for
straight mode. In crossover mode, the FLP appears on
pins 3-6 of RJ-45 and in straight mode, the FLP appears on
pins 1-2.
A: We have received a number of questions regarding
inter-operability of National’s DP83865 with Broadcom’s
BCM5400 1000/100 Mbps PHY. National’s DP83865 is
compliant to IEEE 802.3ab and it is also inter-operable with
the BCM5400 as well as other Gigabit Physical Layer prod-
ucts. However, there are certain situations that might
require extra attention when inter-operating with the
BCM5400.
7.11 I have forced 10 Mbps or 100 Mbps operation
but the associated speed LED doesn’t come on.
A: Speed LEDs are actually an AND function of the speed
and link status. Regardless of whether the speed is forced
or Auto-Negotiated, there has to be good link for the speed
LEDs to turn on.
There are two types of BCM5400’s, those with silicon revi-
sions earlier than C5 and those with silicon revisions of C5
and older. There is a fundamental problem with earlier sili-
con revisions of the BCM 5400, whereby the part was
designed with faulty start-up conditions (wrong polynomials
were used) which prevented the Broadcom BCM5400 from
ever linking to an IEEE 802.3ab compliant part.
7.12 I know I have good link, but register 0x01, bit
2 “Link Status” doesn’t contain value ‘1’ indicat-
ing good link.
This problem was observed in early inter-operability testing
at National Semiconductor. A solution was put together that
allows the DP83865 to inter-operate with any IEEE
802.3ab compliant Gigabit PHY as well as with earlier revi-
sions of the BCM5400 that are non compliant. To enter into
this mode of operation you can either pull pin 1
(NON_IEEE_STRAP) high through a 2kΩ resistor or write
‘1’ to bit 9 of register 0x12.
A: This bit is defined by IEEE 802.3u Clause 22. It indi-
cates if the link was lost since the last time this register was
read. Its name (given by IEEE) is perhaps misleading. A
more accurate name would have been the “Link lost” bit. If
the actual present link status is desired, then either this
register should be read twice, or register 0x11 bit 2 should
be read. Register 0x11 shows the actual status of link,
speed, and duplex regardless of what was advertised or
what has happened in the interim.
7.16 How do I quickly determine the quality of the
link over the cable ?
7.13 Your reference design shows pull-up or pull-
down resistors attached to certain pins, which
conflict with the pull-up or pull-down information
specified in the datasheet?
A: Idle error indicates either that the cable length is beyond
the specified limit or the cable plant does not meet the EIA
568 Category V requirements. The Activity LED indicates
the occurrence of idle error or packet transfer. You monitor
the quality of the link by viewing the Activity LED during
idle.
A: The pull-up or pull-down information specified in the pin
description section of the datasheet, indicate if there is an
internal pull-up or pull-down resistor at the IO buffer used
for that specific pin. These resistors are between 25 - 80
kΩ. They will determine the default strap value when the
pin is floating. If the default value is desired to be changed,
an external 2 kΩ pull-up or pull-down resistor can be used.
7.17 What is the power up sequence for
DP83865?
A: The DP83865 has two types of power supplies, core
and I/O. Although there has not been revealing of power
up sequence error such as latch up or dead lock, it is rec-
ommended that core power takes precedence over the I/O
power when powering up. 1.8V should be up before 2.5V
and 3.3V. When powering down, I/O takes precedence
over core. 2.5V and 3.3V should be turned off before 1.8V.
7.14 How is the maximum package case temper-
ature calculated?
A: The maximum die temperature is calculated using the
following equations:
TJ = TA + Pd(ΟJA
)
TJ = TC + Pd(Οc)
83
www.national.com
7.0 Frequently Asked Questions (Continued)
7.18 What are some other applicable documents?
A: For updated collateral material, please go to “solu-
tions.national.com” website.
— DP83865 Reference Design (Demo board, Schematics,
BOM, Gerber files.)
— Application Note 1263 “DP83865 Gig PHYTER V
10/100/1000 Ethernet Physical Layer Design Guide”
— Application Note 1337 “Design Migration from DP83861
to DP83865”
— Application Note 1301 “Dual Foot Print Layout Notes for
DP83865 Gig PHYTER V and DP83847 DS PHYTER II”
— Application Note 1329 “DP83865 and DP83864 Gigabit
Physical Layer Device Trouble Shooting Guide”
— IEEE 802.3z “MAC Parameters, Physical Layer, Repeat-
er and Management Parameters for 1000 Mbps Opera-
tion.”
— IEEE 802.3ab “Physical layer specification for 1000
Mbps operation on four pairs of category 5 or better bal-
anced twisted pair cable (1000BASE-T)“.
— IEEE 802.3 and 802.3u (For 10/100 Mbps operation.)
www.national.com
84
NOTES
www.national.com
85
8.0 Physical Dimensions inches (millimeters) unless otherwise noted
128 Lead Plastic Flat Pack
LIFE SUPPORT POLICY
NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT
DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT AND GENERAL
COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein:
1. Life support devices or systems are devices or systems which, 2. A critical component is any component of a life support device
(a) are intended for surgical implant into the body, or (b) support
or sustain life, and whose failure to perform, when properly used
in accordance with instructions for use provided in the labeling,
can be reasonably expected to result in a significant injury to the
user.
or system whose failure to perform can be reasonably expected
to cause the failure of the life support device or system, or to af-
fect its safety or effectiveness.
BANNED SUBSTANCE COMPLIANCE
National Semiconductor certifies that the products and packing materials meet the provisions of the Customer Products Stewardship
Specification (CSP-9-111C2) and the Banned Substances and Materials of Interest Specification (CSP-9-111S2) and contain no “Banned
Substances” as defined in CSP-9-111S2.
National Semiconductor
Corporation
National Semiconductor
Europe
National Semiconductor
Asia Pacific
National Semiconductor
Japan Ltd.
Tel: 1-800-272-9959
Fax: 1-800-737-7018
Email: support@nsc.com
Customer Response Group
Tel: 65-254-4466
Fax: 65-250-4466
Tel: 81-3-5639-7560
Fax: 81-3-5639-7507
Fax:
+49 (0) 180-530 85 86
Email: europe.support@nsc.com
Deutsch Tel:
+49 (0) 69 9508 6208
+44 (0) 870 24 0 2171
+33 (0) 1 41 91 8790
English Tel:
Email: ap.support@nsc.com
Francais Tel:
www.national.com
National does not assume any responsibility for use of any circuitry described, no circuit patent licenses are implied and National reserves the right at any time without notice to change said circuitry and specifications.
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications, enhancements, improvements,
and other changes to its products and services at any time and to discontinue any product or service without notice. Customers should
obtain the latest relevant information before placing orders and should verify that such information is current and complete. All products are
sold subject to TI’s terms and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in accordance with TI’s standard
warranty. Testing and other quality control techniques are used to the extent TI deems necessary to support this warranty. Except where
mandated by government requirements, testing of all parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for their products and
applications using TI components. To minimize the risks associated with customer products and applications, customers should provide
adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right, copyright, mask work right,
or other TI intellectual property right relating to any combination, machine, or process in which TI products or services are used. Information
published by TI regarding third-party products or services does not constitute a license from TI to use such products or services or a
warranty or endorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual
property of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of TI information in TI data books or data sheets is permissible only if reproduction is without alteration and is accompanied
by all associated warranties, conditions, limitations, and notices. Reproduction of this information with alteration is an unfair and deceptive
business practice. TI is not responsible or liable for such altered documentation. Information of third parties may be subject to additional
restrictions.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that product or service voids all
express and any implied warranties for the associated TI product or service and is an unfair and deceptive business practice. TI is not
responsible or liable for any such statements.
TI products are not authorized for use in safety-critical applications (such as life support) where a failure of the TI product would reasonably
be expected to cause severe personal injury or death, unless officers of the parties have executed an agreement specifically governing
such use. Buyers represent that they have all necessary expertise in the safety and regulatory ramifications of their applications, and
acknowledge and agree that they are solely responsible for all legal, regulatory and safety-related requirements concerning their products
and any use of TI products in such safety-critical applications, notwithstanding any applications-related information or support that may be
provided by TI. Further, Buyers must fully indemnify TI and its representatives against any damages arising out of the use of TI products in
such safety-critical applications.
TI products are neither designed nor intended for use in military/aerospace applications or environments unless the TI products are
specifically designated by TI as military-grade or "enhanced plastic." Only products designated by TI as military-grade meet military
specifications. Buyers acknowledge and agree that any such use of TI products which TI has not designated as military-grade is solely at
the Buyer's risk, and that they are solely responsible for compliance with all legal and regulatory requirements in connection with such use.
TI products are neither designed nor intended for use in automotive applications or environments unless the specific TI products are
designated by TI as compliant with ISO/TS 16949 requirements. Buyers acknowledge and agree that, if they use any non-designated
products in automotive applications, TI will not be responsible for any failure to meet such requirements.
Following are URLs where you can obtain information on other Texas Instruments products and application solutions:
Products
Audio
Applications
www.ti.com/audio
amplifier.ti.com
dataconverter.ti.com
www.dlp.com
Communications and Telecom www.ti.com/communications
Amplifiers
Data Converters
DLP® Products
DSP
Computers and Peripherals
Consumer Electronics
Energy and Lighting
Industrial
www.ti.com/computers
www.ti.com/consumer-apps
www.ti.com/energy
dsp.ti.com
www.ti.com/industrial
www.ti.com/medical
www.ti.com/security
Clocks and Timers
Interface
www.ti.com/clocks
interface.ti.com
logic.ti.com
Medical
Security
Logic
Space, Avionics and Defense www.ti.com/space-avionics-defense
Transportation and Automotive www.ti.com/automotive
Power Mgmt
Microcontrollers
RFID
power.ti.com
microcontroller.ti.com
www.ti-rfid.com
Video and Imaging
www.ti.com/video
OMAP Mobile Processors www.ti.com/omap
Wireless Connectivity www.ti.com/wirelessconnectivity
TI E2E Community Home Page
e2e.ti.com
Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265
Copyright © 2011, Texas Instruments Incorporated
相关型号:
©2020 ICPDF网 联系我们和版权申明