TI TPS2383PM

SLUS559B − APRIL 2003 − REVISED JULY 2004
D Load Overcurrent and Undercurrent
FEATURES
D Compliant to Power Over Ethernet IEEE
D
D
D
D
D
D
D
D
802.3af Standard
Two-Point 25-kΩ Resistor Discovery
Capacitive Detection for Non-Compliant
Legacy Loads
Power Classification
Controlled di/dt Ramp Power-Up and
Power-Down for EMI Reduction
Current Management for Charging Powered
Device Bulk Capacitance
Electronic Circuit Breaker
Fault detection
Input Undervoltage Lockout (UVLO)
D
D
D
D
D
D
D
D
Detection
12-Bit Port Current and Voltage Acquisition
Standard Slave I2C Serial Interface
5-Bit Serial Address Selectability
Discovery and/or Classification Bypass
Modes Selectable Via Register
Opto-Coupler Compatible SDA and SCL
Lines for System Ground Isolation
Dual Color LED Driver for Port Status
Hardware FAULT Interrupt
Single +48-V Supply Capable
APPLICATIONS
D PoE Switches
D Mid-Span Injectors
APPLICATION DIAGRAM
SWITCH/HUB
CT Choke
POWERED DTR
RJ−45
w/grn
1
TX
2
CT Choke
3
RX
6
4
RJ−45
w/grn
1
grn
grn
w/org
w/org
org
org
blue
blue
CT Choke
RX
2
3
CT Choke
TX
6
4
PPTC Fuse
TPS2383
PSE P
(1 Port)
N
V48
GND
5
7
w/blue
w/blue
w/brn
w/brn
brn
G
8
RS
568A
brn
5
7
PD
Signature
8
PD
DC/DC
Supply
568A
Up to 350 feet of category 5 cable
RG
Optically Coupled
I2C Serial Bus
+
48−V
Supply
−
MSP430
Controller
UDG−03061
! "#$ %!&
% "! "! '! ! !( !
%% )*& % "!+ %! !!$* $%!
!+ $$ "!!&
Copyright  1999 − 2004, Texas Instruments Incorporated
www.ti.com
1
SLUS559B − APRIL 2003 − REVISED JULY 2004
DESCRIPTION
The TPS2383 family of products are power sourcing equipment power managers (PSEPM) that are compliant
to the power-over-ethernet (PoE) IEEE 802.3af Standard. A PSEPM port can discover, classify and deliver
power to a powered device (PD) capable of accepting PoE twisted pair cable. The TPS2383 is fully
programmable by the user. This allows for the detection and powering of both fully compliant 802.3 devices as
well as custom detection of legacy devices.
The TPS2383 PSEPM can individually manage power for up to eight ethernet ports. All operations of the
TPS2383 are controlled through register read and write operations over a standard (slave) I2C serial interface.
The TPS2383 has dual SDA lines to allow easy application of opto-coupler circuitry to maintain ethernet port
isolation when a ground based controller is used. Each TPS2383 has five selection pins making it possible to
address up to 32 devices on the I2C bus and allows individual control and monitoring of up to 256 ethernet ports
from a single master I2C controller. Per-port write registers initiate and manipulate the flow of the discovery,
classification, and power-up states while the read registers contain status information of the enable process,
faults, classification value, and real time port operating current and voltage. Per-port status LED drivers are
provided which can be manually controlled through the serial I/O.
The TPS2383 is available in a full function 64-pin LQFP.
External N-channel MOSFETs switch port power. High-voltage (HV) gate drive ensures that these FETs are fully
enhanced, resulting in lower power dissipation and enabling the use of lower costs FETs. The TPS2383
generates its HV gate supply from the 48-V port power, simplifying system power supply design. An external
3.3-V digital supply is also used. This supply can be active when 48-V power is not present which allows the
user to access the part through the serial I/O in this case. A 5-V analog supply is used to power port LEDs and
internal analog functions. Due to the very low quiescent current, both the 3.3-V and 5-V supply can be generated
from the +48-V power bus with minimal external components. An internal power-on-reset (POR) circuit with an
ORed external input pin resets all registers positions to a known safe state upon power up.
2
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
These devices have limited built-in ESD protection. The leads should be shorted together or the device placed in conductive foam
during storage or handling to prevent electrostatic damage to the MOS gates.
ABSOLUTE MAXIMUM RATINGS
over operating free-air temperature range unless otherwise noted(1)
TPS2383
Input voltage range, VCC
UNIT
V48, 1P, 2P, 3P, 4P, 5P, 6P, 7P, 8P, 1N, 2N, 3N, 4N, 5N, 6N, 7N, 8N
−0.5 to 65
1RS, 2RS, 3RS, 4RS, 5RS, 6RS, 7RS, 8RS, 1G, 2G, 3G, 4G, 5G,
6G, 7G, 8G
−0.5 to 12
VL
−0.5 to 3.9
V5
−0.5 to 6
SCL_I, SDA_I, SDA_O, INTB, A1, A2, A3, A4, A5, EN, PORB
V
−0.5 to 6
Storage temperature, Tstg
−55 to 150
Operating temperature, TJ
0 to 70
°C
C
Lead temperature, Tsol, 1,6 mm (1/16 inch) from case for 10 seconds
260
(1) Stresses beyond those listed under “absolute maximum ratings” may cause permanent damage to the device. These are stress ratings only,
and functional operation of the device at these or any other conditions beyond those indicated under “recommended operating conditions” is
not implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability. All voltages are with respect to
GND. Currents are positive into and negative out of, the specified terminal.
RECOMMENDED OPERATING CONDITIONS
MIN
Input voltage, VV48
44
Operating junction temperature, TJ
0
NOM
MAX
UNIT
48
57
V
70
°C
MAX
UNIT
ELECTROSTATIC DISCHARGE (ESD) PROTECTION
Human body model
1.5
CDM
1
Machine model
kV
0.2
ORDERING INFORMATION
PACKAGED DEVICES(2)
TA
LQFP−64
(PM)
0°C to 70°C
TPS2383PM
(2) The PM package is available taped and reeled. Add R suffix to
device type (e.g.TPS2383PMR) to order quantities of 2,500
devices per reel.
www.ti.com
3
SLUS559B − APRIL 2003 − REVISED JULY 2004
ELECTRICAL CHARACTERISTICS
VV48= 48 V, RT = 120 kΩ, 0°C to 70°C, and TA = TJ (unless otherwise noted)
TEST CONDITIONS
PARAMETER
MIN
TYP
MAX
UNIT
POWER SUPPLIES
Quiescent current, V48
Off mode (all ports)
3
5
Powered mode (all ports)
5
10
3
15
Quiescent current, V5
Powered mode (all ports),
Quiescent current, VL
Powered mode (all ports),
VV5 = 5 V
VVL = 3.3 V
Internal analog supply, V10
Internal reference(1)
Input UVLO
Internal POR timeout (I2C)
After all supplies are good (8 CLK periods)
Internal POR timeout (state machine)
After all supplies are good (65536 CLK periods)
0.28
4
9.75
10.50
11.50
2.475
2.500
2.525
28
37
40
mA
V
8
CLKs
65536
LED OUTPUTS
High-level output voltage, L1 through L8
Low-level output voltage, L1 through L8
Tri−state leakage(1)
ISRC = 5 mA
ISINK = 5 mA
4
0.75
V
µA
0.1
ANALOG CIRCUITS − PORT VOLTAGE CONTROL LOOP
Input resistance, nP to nN
240
400
15.0
17.5
Discovery voltage, high
7.5
Discovery voltage, low
3.5
Classification voltage
A/B select = B
Loop power supply feedthru loop control
range C(1)
640
kΩ
8.8
9.5
V
4.4
5.5
1.5
Discovery short-circuit current
mV
3
Port output, undervoltage
40
42
44
Port output, overvoltage
51
53
55
Classification current limit
ClassLimit1 = 0,
ClassLimit2 = 0
160
ClassLimit1 = 1,
ClassLimit2 = 0
80
ClassLimit1 = 0,
ClassLimit2 = 1
40
ClassLimit1 = 1,
ClassLimit2 = 1
20
mA
V
mA
N-CHANNEL MOSFET GATE CONTROL
Gate turn-off MOSFET RDS(on)
Maximum gate voltage
8
10
Gate turn-off delay time from overcurrent
fault
Gate turn-off timer from UV/OV fault
After port enabled and ramped up,
(1024 CLK periods)
UV/OV spike timer, power quality warning(1)
(256 CLK periods)
Gate turn-off timer from overload fault
(32768 CLK periods)
NOTE: (1) Ensured by design. Not production tested.
4
Ω
200
www.ti.com
12
V
7.5
µs
1024
256
32768
CLKs
SLUS559B − APRIL 2003 − REVISED JULY 2004
ELECTRICAL CHARACTERISTICS
VV48= 48 V, RT = 120 kΩ, 0°C to 70°C, and TA = TJ (unless otherwise noted)
TEST CONDITIONS
MIN
TYP
MAX
Overload threshold voltage
165
187
207
Peak current threshold voltage
175
213
250
50
100
Reset voltage, CINT
10
100
Maximum swing, CR
4.5
6.0
Input leakage, nRS
10
PARAMETER
UNIT
LOW-SIDE CURRENT-SENSE
Overload timout
Maximum swing, CINT
ANALOG-TO-DIGITAL CONVERTER
A/D resistive scaling(1)
9
RRD = 1 kΩ
11
Referenced to port #1
RRS = 0.5 Ω
35
A/D load current scaling(1)
A/D load voltage scaling(1)
RRS = 0.5 Ω
4.72
−5%
0%
ILOAD = 50 mA,
ILOAD = 300 mA,
A/D port voltage conversion
VPORT = 5 V
VPORT = 45 V
RRS = 0.5 Ω
236
RRS = 0.5 Ω
1416
V
mV
V
count/kΩ
5%
count/mA
33.6
A/D load current conversion
ms
µA
72
Port-to-port resistance variation
A/D classificatin scaling(1)
mV
count/V
count
168
1512
DIGITAL I/O
Logic input threshold voltage, SCL,
SDA_I, A1 through A5, EN, PORB
1.5
Input hysteresis, SCL, SDA_I
250
Input hysteresis, EN, PORB
150
Input pull-down resistance, EN, PORB
V
mV
50
Pull-up current, A0, A1, A2, A3, A4, A5
10
kΩ
25
Drain = 6 V
Logic low output voltage, SDA_O
ISINK = 3 mA
Drain = 6 V
200
10
µA
ISINK = 3 mA
200
mV
Logic high leakage, INTB
Logic low output voltage, INTB
DIGITAL I/O TIMING I2C CHARACTERISTICS
10
µA
A
Logic high leakage, SDA_O
Clock frequency, SCL
Pulse duration
0
SCL high
1.3
SCL low
0.6
Rise time, SCL, SDA(1)
Fall time, SCL, SDA(1)
mV
400
kHz
µss
300
300
Setup time SDA to SCL
250
Hold time SCL to SDA
ns
0
Bus free time between startup and stop
1.3
Setup time SCL to start condition
0.6
Hold time start condition to SCL
0.6
Setup time, SCL to stop condition
0.6
µss
NOTE: (1) Ensured by design. Not production tested.
www.ti.com
5
SLUS559B − APRIL 2003 − REVISED JULY 2004
TERMINAL FUNCTIONS
TERMINAL
NAME
PIN
I/O
DESCRIPTION
POWER AND GROUND
AG1
26
I
Analog ground 1. Analog ground of the V5, V10 and V48 power systems. It should be externally tied to the
common copper 48-V return plane. This pin should carry the low side of two de-coupling capacitors tied to V48
and V10
AG2
51
I
Analog ground 2. Analog ground, which ties to the substrate and ESD of the device. It should be externally tied
to the common copper 48-V return plane. AG1 and AG2 must be tied together directly for lowest noise
operation.
DG
56
I
Digital ground. It connects to the internal logic ground bus. It should be externally tied to the common copper
48V return plane. In addition a 0.1uF de−coupling capacitor should terminate as close to this node and the VL
pin as possible.
RG
29
I
Reference ground. A precision sense of the external ground plane. It should also be used as the ground guard
ring for the integration capacitor (CINT). It should be the closest ground to the low side of the 0.5-Ω current
sense resistors, as well as RD, CINT, and RT. It should tie to common copper 48-V return plane.
V10
28
O
+10V analog supply.Connects to the internal analog power bus. This voltage is generated internally. This pin
should not be tied to any external supplies. A 0.1-µF de-coupling capacitor should terminate as close to this
node and the AG2 pin as possible. This pin can be used for external generation of V5.
V48
27
I
+48V input to the device. This supply can have a range of 44 V to 57 V. This pin should be de-coupled with a
0.1-µF capacitor from V48 to AG2 placed as close to the device as possible.
V5
25
I
External +5V analog supply.Connects to the internal analog power bus. This supplies the LED output drivers
and internal analog circuits. A 0.1-µF de-coupling capacitor should terminate as close to this node and the AG1
pin as possible.
VL
54
I
External +3.3V logic supply. This pin connects to the internal logic power bus. This is the supply voltage for the
internal device logic. A 0.1-µF de-coupling capacitor should terminate as close to this node and the DG pin as
possible. This pin can be powered from V5 by using a 4.22-kΩ resistor from V5 to VL..
PORT ANALOG SIGNALS
6
1P
1
I
2P
8
I
3P
9
I
4P
16
I
5P
33
I
6P
40
I
7P
41
I
8P
48
I
1N
2
I
2N
7
I
3N
10
I
4N
15
I
5N
34
I
6N
39
I
7N
42
I
8N
47
I
Port positive. +48V load sense pin. Terminal voltage is monitored and controlled differentially with respect to nN.
Optionally, if the application warrants it, this high-side path can be protected with the use of a self-resetting poly
fuse.
Port negative. −48V load return sense pin. The low side of the load is switched and protected with the external
N-channel MOSFET.
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
TERMINAL FUNCTIONS (continued)
TERMINAL
NAME
PIN
I/O
DESCRIPTION
PORT ANALOG SIGNALS (continued)
1G
3
O
2G
6
O
3G
11
O
4G
14
O
5G
35
O
6G
38
O
7G
43
O
8G
46
O
1RS
4
I
2RS
5
I
3RS
12
I
4RS
13
I
5RS
36
I
6RS
37
I
7RS
44
I
8RS
45
I
Port gate. Connect to the gate of an external N-channel MOSFET. During turn-on, this pin is controlled by a
linear current amplifier (LCA) such that the load current ramps up from zero to a maximum sourcing current of
425 mA. This pin is driven to as high as 10 V. During controlled turn-off, this pin is driven such that the load
current ramps down from a maximum of 425 mA to zero. The capacitor on the CR pin is utilized to generate the
ramp control signal voltages. During a fault turn-off this pin is discharged quickly with a low on-resistance
internal switch.
Port resistor sense. This is the kelvin sense path for the high potential end of the load current sense resistor.
Parameters controlled by the load current sense resistor include: the average undercurrent/overcurrent and
peak-load current thresholds, the peak PD inrush current limit during startup, and the nominal classification
current levels. Use a 0.5-Ω load current sense resistor to be compliant to the 802.3 specification levels.
ANALOG SIGNALS
CR
50
I
Ramp capacitor. During load power up and down, this capacitor is used as the di/dt current slew capacitor. A
1.5-V peak triangular waveform is present on this pin during ramp up/down. Connect a 0.1-µF capacitor from
this pin to AG2 and a 120-kΩ resistor at RT to meet the 802.3af specification timing levels.
CT
52
I
Timing capacitor. This capacitor and the resistor on the RT pin set the internal clock frequency of the device.
This clock is used for the internal state machine, integrating A/D counters, POR time-out, and fault and delay
timers of each port. Use a 100-pF to 470-pF capacitor for CT and a 120-kΩ resistor on RT to set the internal
clock in a range of 100 kHz to 500 kHz. This timing can be overridden by driving the CT pin with a 0 V to 5 V
square wave with a frequency from 0 kHz to 500 kHz.
CINT
30
I
This capacitor is used for the ramp A/D converter signal integration. Connect a 0.033-µF capacitor from this pin
to RG. For minimal errors due to dielectric absorption, use a poly or Teflon capacitor type. Ceramic types can be
used, but note the increased conversion error.
L1
17
O
L2
18
O
L3
19
O
L4
20
O
L5
21
O
L6
22
O
L7
23
O
L8
24
O
RD
32
I
The discovery current-sense resistor is connected in the path from the RD pin to RG ground. The discovery
current-sense resistor sets the discovery value to 25-kΩ (nominal) when a 665-Ω value is used. For best noise
performance, de-couple this pin with a 0.68-µF, ceramic capacitor to RG ground.
RT
31
I
Bias set resistor. This resistor sets all precision bias currents within the device. This pin is forced to an internal
1.25-V reference voltage level. The current that flows into this resistor due to the applied 1.25-V bias is
replicated and used throughout the device. This resistor also works in conjunction with the capacitors on CR,
CT and CINT to set internal timing values. Use a 120-kΩ resistor to be compliant to the requirements of 802.3af.
LED lamp drivers. Dual or single color LEDs can be connected to each of these pins. Each pin indicates the
state of the corresponding port. This is a tri-state port that is under full control of the host micro-controller. As
such it can also be used as a data port, or general-purpose output driver.
www.ti.com
7
SLUS559B − APRIL 2003 − REVISED JULY 2004
TERMINAL FUNCTIONS (continued)
TERMINAL
NAME
PIN
I/O
DESCRIPTION
DIGITAL SIGNALS
A1
64
I
A2
63
I
A3
62
I
A4
61
I
A5
60
I
EN
53
I
This pin is normally to be held low. It has been reserved for future expansion of the TPS238x device family
INTB
59
O
PORB
49
I
This is an open drain output that goes low if a fault condition is produced on any of the eight ports.
This pin can be used to override the internal POR. When held low, the I2C interface, all the state machines, and
registers are held in reset. When all internal and external supplies are within specification, and this pin is set to
a logic high level, the POR delay begins. The I2C interface and registers becomes active within eight CLK
periods of this event and communications to read or preset registers can begin. The reset delay for the
remainder of the device then extinguishes within 65536 CLK periods.
SCL_I
55
I
SDA_I
58
I
SDA_O
57
O
Addresses 1 through 5. This is the I2C address select input. Select the appropriate binary address on these pins
by connecting this pin to device ground for a logic low and tying this pin to the VL pin for a logic high.
Serial clock input pin for the I2C interface.
Serial data input pin for the I2C interface. When jumpered with the SDA_O pin, this connection becomes the
standard bi-directional serial data line (SDA).
Serial data open drain output for the I2C interface. When jumpered with the SDA_I pin, this connection
becomes the standard bi-directional serial data line (SDA). This is a high-voltage open drain output that can
drive opto-coupler LEDs directly from the +48-V bus with an external, series current limiting resistor.
PACKAGE DESCRIPTION
1
64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49
48
2
47
3
46
4
45
5
44
6
43
7
42
8
41
9
40
10
39
11
38
12
37
13
36
14
35
15
34
33
16
1718 19 20 21 22 23 24 25 26 27 28 29 30 31 32
L1
L2
L3
L4
L5
L6
L7
L8
V5
AG1
V48
V10
RG
CINT
RT
RD
1P
1N
1G
1RS
2RS
2G
2N
2P
3P
3N
3G
3RS
4RS
4G
4N
4P
CR
PORB
A1
A2
A3
A4
A5
INTB
SDA_I
SDA_O
DG
SCL_I
VL
EN
CT
AG2
PM PACKAGE
(TOP VIEW)
8
www.ti.com
8P
8N
8G
8RS
7RS
7G
7N
7P
6P
6N
6G
6RS
5RS
5G
5N
5P
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
The TPS2383 architecture has been designed to work efficiently with simple low-cost controllers such as those
in the MSP430 family of devices. Reference design and code examples for complete PSE management are
available from TI for the TPS2383/MSP430 chipset solution.
The PSEPM discovery method, as defined in the IEEE 802.3af Standard, uniquely identifies a powered device
25-kΩ resistor. Use of low-level probe signals during discovery prevents damage to non-802.3 devices. The use
of a point-to-point slope detection method for the PD 25-kΩ resistor measurement allows accurate detection,
even if series steering diodes are present at the PD. For legacy loads, capacitive detection can be enabled. In
this mode the TPS2383 A/D is used to measure the loads capacitive value.
After a successful discovery of the PD, the TPS2383 has a classification feature to identify the expected PD
power level based on a current signature from the PD. The classification current level is measured at a reduced
terminal voltage of 17.5 V and classified with 12 bits of resolution. The controller can then use this information
to classify per the levels of the IEEE 802.3af standard or use levels custom to the application. Knowledge of
the expected load power allows the power sourcing equipment to be built with a smaller and less expensive
system power supply. For installations where classification is not needed, and reduced power-up time is
desired, classification can be bypassed by setting the appropriate bits in the per-port write register.
In classification and powered modes, the PSEPM drives an external low side N-channel MOSFET for control
of the 48-V return line. The use of an external N-channel MOSFET enables selection of very low RDS(on) devices
to minimize board power dissipation in enclosures that may be controlling 100’s of ports. Current sensing is
performed with a low value resistor, again minimizing board power dissipation. In discovery mode, due to the
very low current used, an internal N-channel MOSFET is utilized in conjunction with an external, high value,
current sense resistor.
The TPS2383 identifies all fault conditions defined in the PoE IEEE 802.3af Standard. The monitored conditions
include input undervoltage lockout (UVLO), output undervoltage (UV) and overvoltage (OV), average and peak
overcurrent detection, average undercurrent detection, and run current. If a fault condition is detected during
power ramp or at any other time, the PSEPM circuit breaks by disconnecting the 48-V return line and then
updates the fault status of the corresponding port register. When the TPS2383 is disabled the PSEPM ramps
the current down at a controlled rate and the PSEPM changes states to a lower power sleep mode.
To maintain full compliance to Underwriters Laboratory (UL), IEC950 SELV and NEMKO safety standards an
optional, low cost self-resetting PTC fuse can be added to the unswitched +48-V line to protect the system
supply and wiring infrastructure from secondary building wiring faults.
When the PSEPM is enabled and a PD is discovered and optionally classified, power is ramped to the PD at
a controlled current ramp rate to reduce EMI. Upon completion of the current ramp up, the port current remains
limited at less than 400 mA. Upon startup the port can remain in current limit for a timed value of 50 ms which
allows the bulk filter capacitance of the PD to charge. Once a PD is successfully powered and the external
N-channel MOSFET is fully on, the average and peak current to the PD is continuously monitored. A
disconnected load is detected if the average current falls below 10 mA. An overcurrent is detected if the average
current exceeds 375 mA and/or the peak current exceeds 425 mA. If any of these conditions exist, status bits
are set in the per-port read register set and the power is removed from the load.
For maximum rejection of external wiring and power supply noise sources during the measurement of line
current, voltage, resistance or capacitance in discovery, classification and powered modes, a proprietary low
noise A/D converter is used. Converted measurements are processed and compared with digital set-points for
limit compliance. The 12-bit conversion of this data-acquisition system is available through the read register
enabling measurement of the discovery resistance, classification current and powered mode port running
current and voltage. This is a valuable feature in ethernet switch management as it allows monitoring of real-time
parameters across the system network.
The TPS2383 is available in a full function 64 pin LQFP package.
www.ti.com
9
SLUS559B − APRIL 2003 − REVISED JULY 2004
TPS2383 Evaluation Board
The full performance features of the TPS2383/MSP430 chipset can be demonstrated with the
TPS2383EVM−001 evaluation board. The TPS2383EVM−001 is a scaleable system that can be expanded to
support 48 ports using a single microcontroller. This evaluation design can be used as a hardware/firmware
template for modification to specific customer requirements. Please contact Texas Instruments or refer to the
TPS2383EVM−001 Users Guide (SLUU168) for complete information.
FUNCTIONAL BLOCK DIAGRAM
TPS2383
Chip Power Supply
Management
PSE Port Section
PSE PORT 8
PSE PORT 7
PSE PORT 6
PSE PORT 5
PSE PORT 4
PSE PORT 3
PSE PORT 2
PSE PORT 1
V48
ANALOG
SUPPLY
V10
WRITE
REGISTER
POR
PORB
PORT
INTERFACE
4 Pins/Port
(32 PINS)
PORT
PORT
CONTROLLER MONITOR
STATE
and FET
READ REGISTER
MACHINE
CONTROL
INPUT UVLO
INTERNAL ANALOG BUS
V5
INTERNAL DIGITAL BUS
VL
RAMP
STATE
MACHINE
POWER
RAMP
CONTROL
LED DRIVERS
MUX
12 BIT A/D
A/D STATE MACHINE
SCL
VOLTAGE
SDA_I
REFERENCE
2 WIRE SERIAL IO
SLAVE I2C CONTROLLER
MASTER
SEQUENCER
STATE
MACHINE
CHIP BIAS
SDA_O
MAIN OSC and CLK
ANALOG
CONTROL LOGIC
LED
DRIVERS
L1−L8
(8 PINS)
FUTURE
ENHANCEMENT
INTB
DEVICE ADDRESS
and INTERRUPT
A1
A2
A3
A4
A5
PORT ADDRESS PINS A1 − A5
10
EN
CT
GNDS
(2 ANA PINS,
1 DIG PIN)
www.ti.com
RT
RG
CINT
RD
CR
UDG−03062
SLUS559B − APRIL 2003 − REVISED JULY 2004
TPS2383 BLOCK DIAGRAM − SHARED PORT ANALOG RESOURCES
Poly PTC Fuse
BBR550, 90 V, 0.55 A
+48 V
8 to 1 MUX
C1
0.47 µF
N
Overload Timer
NEGOUT
0.187 V
4 to 1 MUX
4 to 1 MUX
P
VUV
Ramp State
Machine
Current
Command
Offset
Correction
+
12-bit
A/D
UV/OV Filter
Port State Machine
Voltage
Control
A/D State
Machine
CINT
0.033µF
+
+
+
8 to 1 MUX
Power to
Ethernet
XFMR
POSOUT
VOV
+
Discovery/
Class
Spike Filter
TPS2383 Port Resources
TPS2383 Shared Resources
A/B
select
A
Linear Current B
Controller
250 µA
+
4 to 1 MUX
2 µA
RS
1.5 V
Rs
0.5
Overcurrent
+
25 µA
G
0.213 V
+
+
7-µs Digital Filter
RG
8:1
1 of 8 channels
50
Fast
Ramp
32
Slow
Ramp
26
RD
CR
CR
0.1 µF
AG1
RD
665 Ω
CD
0.68 µF
51
AG2
48 V
RTN
UDG−03063
www.ti.com
11
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
SEPM STATE MACHINES
The TPS2383 has circuit resources that are common to each port and circuit resources that are shared by all
ports. Five independent state machines are used to control the common and shared PSEPM resources. Port
control, UV/OV/OC and overload protection are all features that are common to each port. Data acquisition and
power ramping are shared functions for all the ports.
The master sequencer state machine is used to index the port presently being serviced and to distribute the
shared resources to the currently selected port. The single master sequencer is responsible for incrementally
accessing ports 1 through 8 and allowing those ports to process register data when they are accessed. Ports
1 through 8 each have a port sequencer, which controls all the power enabling and fault protection functions
of the port per the register commands. The A/D has an A/D Sequencer that triggers, cycles and signals the port
and master sequencer upon completion. The ramp sequencer controls the power ramping resource and is
triggered by the port and master sequencer and provides a completion signal when power ramping is over.
Upon power-up the master sequencer is enabled and running after a POR delay and begins acting on register
commands. A shorter POR delay releases the reset on the I2C function and registers before the port reset is
removed. This arrangement allows for register setup and polling over the I2C bus quickly upon power up but
ensures that power cannot be applied until the power supply is fully energized and stable. The default power-up
state for all command registers is a null condition. The state sequence order of the TPS2383 is discovery,
classification and power delivery if a POE compliant device is detected on the other end of the data cable.
The master sequencer powers-up in a default free-running mode. The TPS2383 also has a JOG mode. By
setting the JOG_MODE register bit high, the master sequencer then no longer runs freely, but increments to
the next sequential port each time the JOG register bit is set to a logic 1. The JOG bit is self-cleared once the
port increments to the next position.
Sequencing starts with port 1 and ends with port 8 and then repeats. The port sequencer signals status
information to the master sequencer and skips over disabled ports. When the master sequencer detects an
enabled port, it pauses at that port until discovery, classification and power-up is complete before proceeding
to the next. When the master sequencer reaches a powered port, it pauses and take a reading of the ports
run-time current and/or voltage before proceeding to the next port. When a powered PD load is disconnected,
the disconnect event can be detected the next time that port is selected by the master sequencer. When the
disable bit of a powered port is set in the corresponding register, that operation is completed the next time the
master sequencer selects that port. An overcurrent fault event shuts down the offending port independent of
any sequencer state.
DUAL COLOR LED DRIVERS
The LED driver pins (L1 through L8) can be used to drive single or dual, color LEDs. These LEDs are intended
to provide installation or service personnel with the necessary information to install and troubleshoot the system
infrastructure. The Ln pins have internal tri-state drivers. These LEDs can be controlled directly from the I2C
registers. The reset state of all the LEDs is tri-state. Cross-conduction logic disables both internal high- and
low-side MOSFETS if an attempt is made to enable both transistors on a given port. These are high current
(10-mA) drivers that can be used for other applications such as the drive of optocouplers or electromechanical
devices, or can just be used as an 8-bit data port.
12
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
I2C INTERFACE
The serial interface used in the TPS2383 is a standard two-wire I2C slave architecture core. The standard
bi-directional SDA lines of the I2C architecture are broken out into an independent input and output path. This
feature simplifies, earth grounded, controller applications that require opto-isolators to keep the 48-V return of
the ethernet power system floating. For applications where opto-isolation is not required, the bi-directional
property of the SDA line can be restored by connecting SDA_I to SDA_O. The SCL line is a unidirectional input
only line as the TPS2383 is always accesses as a slave device and it never controls the bus.
Data transfers that require a data-flow reversal on the SDA line are four-byte operations. This occurs during a
TPS2383 port read access cycle where a slave address byte is sent, followed by a port/register address byte
write. A second slave address byte is sent followed by the data byte read using the port/register setup from the
second byte in the sequence.
Data write transfers to the TPS2383 do not require a data-flow reversal and as such only a three-byte operation
is required. The sequence in this case is to send a slave address byte, followed by a write of the port/register
address followed by a write of the data byte for the addressed port.
The I2C access cycle consists of the following steps 1 through 7 and is also shown in Figure 4.
1. Start sequence (S)
2. Slave address field
3. Read/write
4. Acknowledge
5. Port/register address or data field
6. Acknowledge/not acknowledge
7. Stop sequence (P)
The I2C interface and the port read write registers are held in active reset until input voltage is within specification
and the internal POR timer has timed out.
Start/Stop
The high-to-low transition of SDA while SCL is high defines the start condition. The low-to-high transition of SDA
while SCL is high defines the stop condition. The master device initiates all start and stop conditions.
A first serial packet enclosed within start and stop bits, consists of a seven-bit address field, read/write bit, and
the acknowledge bit. The acknowledge bit is always generated by the device receiving the address or data field.
Five of the seven address bits are used by the TPS2383. The sixth and seventh bit is a placeholder for future
expansion. During a write operation to the TPS2383 from the master device, the data field is eight bits. During
a read operation where the TPS2383 is writing to the master device, the data field is also eight bits.
www.ti.com
13
14
www.ti.com
Figure 1. I2C Read/Write Cycles
SDA_O
SDA_I
Read Cycle
SDA_O
SDA_I
Write Cycle
SDA
SCL
Start
Bit
Start
Bit
Register/Port
Address
Device Address
R/W=0
Ack
Bit
D7 D6 R2 R1 R0 P2 P1 P0
A7 A6 A5 A4 A3 A2 A1 A0
R/W Bit
Register/Port
Address
Clock 0
Device Address
R/W=0
Ack
Bit
R/W Bit
Clock 1
D7 D6 R2 R1 R0 P2 P1 P0
Clock 1
A7 A6 A5 A4 A3 A2 A1 A0
Start
Condition(S)
Start/Stop Sequence
Ack
Bit
Ack
Bit
Start
Bit
R/W=1
Device Address
Ack
Bit
R/W Bit
A7 A6 A5 A4 A3 A2 A1 A0
Data from
Master to
TPS2383
Ack Stop
Bit Bit
Stop
Condition (P)
D7 D6 D5 D4 D3 D2 D1 D0
Clock 1
Ack
Bit
D7 D6 D5 D4 D3 D2 D1 D0
Data from
TPS2383 to
Master
Stop
Bit
SLUS559B − APRIL 2003 − REVISED JULY 2004
UDG−03060
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
Chip Address
The address field of the TPS2383 is eight bits and contains five bits of device address select, a read/write bit,
and two reserved bits per Table 1. The leading two bits are reserved for future port expansion, and must be set
to 0 for address acknowledge. The five device address select bits follow this. These bits are compared against
the hard-wired state of the corresponding, device address select pins (A1 through A5). When the field contents
are equivalent to the pin logic states, the device is addressed. These bits are followed by a least significant bit
(LSB), which is used to set the read or write condition (1 for read and 0 for write). Following a start condition
and an address field, the TPS2383 responds with an acknowledgement by pulling the SDA line low during the
ninth clock cycle if the address field is equivalent to the value programmed by the pins. The SDA line remains
a stable low while the ninth clock pulse is high.
Table 1. Address Selection Field
BIT
FUNCTION
A7
Future expansion (set to 0)
A6
Future expansion (set to 0)
A5
Device address. Compared with A5
A4
Device address. Compared with A4
A3
Device address. Compared with A3
A2
Device address. Compared with A2
A1
Device address. Compared with A1
A0
Read/write
Port/Register Cycle
After the chip address cycle, the TPS2383 accepts eight bits of port/register select data as defined in Table 2.
The SCL line high-to-low transition after the eighth data bit then latches the selection of the appropriate internal
register for the follow on data read or write operation. After latching the eight-bit data field, the TPS2383 pulls
the SDA line low for one clock cycle.
Data Write Cycle
For a data write sequence, after the Port/Register address cycle, the TPS2383 accepts the eight bits of data.
The data is latched into the previously selected Write Register, and the TPS2383 generates a data acknowledge
pulse by pulling the SDA line low for one clock cycle. To reset the interface, the host or master subsequently
generates a Stop bit by releasing the SDA line during the clock-high portion of an SCL pulse.
www.ti.com
15
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
Data Read Cycle
For a data read sequence, after the register acknowledge bit, the master device generates a Stop condition.
This is followed by a second Start condition, and retransmitting the device address as described in Chip Address
above. For this cycle, however, the R/W bit is set to a 1 to signal the read operation. The TPS2383 again
responds with an acknowledge pulse. The address acknowledge is then followed by sequentially presenting
each of the eight data bits on the SDA line (MSB first), to be read by the host device on the rising edges of SCL.
After eight bits are transmitted, the host acknowledges by pulling the SDA line high for one clock pulse. The
completed data transfer is terminated with the host generating a Stop condition.
Table 2. Register/Port Addressing
BIT
FUNCTION
D7
Future expansion
D6
Future expansion
D5
R2, register select MSB
D4
R1, register select
D3
R0, register select LSB
D2
P2, port address MSB
D1
P1, port address
D0
P0, port address LSB
STATE
000 =
001 =
010 =
011 =
100 =
101 =
Control (common Write register)
Port Status (per port Read register)
Port Control (per port Write register)
A2D Register low byte (common read register)
A2D Register high byte (common read register)
Chip identification/revision (common read register)
000 =
001 =
010 =
011 =
100 =
101 =
110 =
111 =
Port 1
Port 2
Port 3
Port 4
Port 5
Port 6
Port 7
Port 8
Port/Register Cycle
Table 3. Common Control
Write Register, Register Address = 000 (Common Register)
BIT
16
FUNCTION
STATE
PRESET
STATE
D7
Jog_mode
0 = Normal mode
1 = Jog mode
0
D6
Jog
0 = Don’t jog
1 = Jog. This bit is self-clearing. It does not need to be reset to a 0 for each new jog.
0
D5
Bypass ramp
0 = Normal mode
1 = Bypass power−up ramp and powered mode for all ports
0
D4
Disconnect disable
0 = Normal mode
1 = Disable the effect of the logic signal from the Disconnect detection circuits. This is
an expansion function for future parts. This bit should be set to a logic 1
0
D3
Bypass discovery
0 = Normal mode
1 = Bypass discovery mode.
0
D2
Bypass sample
0 = Normal mode
1 = Bypass current sample of all powered ports.
0
D1
Bypass classification
0 = Normal mode
1 = Bypass classification of all ports.
0
D0
Discovery fault disable
0 = Normal mode
1 = Disable the effect of the logic signal from the Discovery circuits. This is
an expansion function for future parts. This bit should be set to a logic 1.
0
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
Port Status Registers
The Port Status registers (address 001) contain port−specific information regarding the operational or faulted
state of each of the eight Ethernet ports, as shown in Tables 4 and 5. In addition, the Port 1 Status register
(Register/Port address xx00/1000) bits D6:D3 contain the current usage information of some of the common
hardware resources, as they are accessed in the servicing of the port presently selected by the master
sequencer.
Table 4. Port Status Port 1
Read Register, Register Address = 001, Port Address = 000
BIT
FUNCTION
STATE
PRESET
STATE
D7
Port service
0 = Port not selected
1 = Port selected and being serviced
0
D6
Down
0 = Selected port is being ramped up
1 = Selected port is not being ramped up
1
D5
Ramp ServiceB
0 = Selected port is using the RAMP module
1 = Selected port is not using the RAMP module
1
D4
Current SampleB
0 = Selected port current is being acquired
1 = Selected port current is not being acquired
1
D3
A to D ServiceB
0 = Discovery/classification data acquisition in process
1 = A/D not performing discovery/classification
1
D2
Fault status (MSB)
D1
Fault status
D0
Fault status (LSB)
000 = Reset state
001 = UV/OV fault/spike
010 = UV/OV spike
011 = Peak over−current fault
100 = Overload fault
101 = Discovery Fail
110 = Load disconnect
111 = Reserved for future
www.ti.com
000
17
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
Table 5. Port Status
Read Register, Register Address = 001, Port Address = 001−111 (Port 2 through Port 8)
BIT
FUNCTION
PRESET
STATE
STATE
0 = Port not selected
1 = Port selected and being serviced
D7
Port service
0
D6
Spare
0
D5
Spare
1
D4
Spare
1
D3
Spare
1
D2
Fault status (MSB)
D1
Fault status
D0
Fault status (LSB)
000 = Reset state
001 = UV/OV fault/spike
010 = UV/OV spike
011 = Peak over−current fault
100 = Overload fault
101 = Discovery Fail
110 = Load disconnect
111 = Reserved for future
000
Port Control Registers
The Port Control registers’ (address 010) bit maps are shown in the following tables. These registers contain
port-specific control bits, accessible over the I2C bus, for setting the port powered and port LED modes. To
conserve hardware resources, some common control functions have been assigned to available bit locations
within these registers. To clarify usage, these common functions are identified as such in the table RANK
column.
Table 6. Port Control
Write Register, Register Address = 010, Port Address = 000 (Port 1 Register)
BIT
FUNCTION
D7
Port fault disable
D6
POR disable
D5
Software RESET
RANK
PRESET
STATE
0 = Normal mode
1 = Disable the port overload and UV/OV timers
0
Common
0 = Normal POR timing
1 = Force POR to a non-reset state
0
Common
0 = Normal operation
1 = Reset all circuits and start a device POR timing cycle
0
0
Port
D4
LED blink enable
Port
0 = LED is continuous
1 = Blink the enabled LED at a fast-blink rate. Note fast-blink rate (in
milliseconds) equivalent to 0.00013 x TPS2383 clock (CLK). This could be
as fast as 65 ms for a 500-kHz, TPS2383 clock.
D3
LED low-side enable
Port
1 = Enable the low-side FET and drive the LED pin low
0
D2
LED high-side enable
Port
1 = Enable the high-side FET and drive the LED pin to V5
0
Port
00 = Port OFF or disable
01 = Discovery − classification − power on sequence
10 = Sample powered-mode current
11 = Power-down an active port
0
D1
Enable modes
D0
18
STATE
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
Table 7. Port Control
Write Register, Register Address = 010, Port Address = 001 (Port 2 Register)
BIT
FUNCTION
D7
Port fault disable
D6
Spare
RANK
Port
STATE
0 = Normal mode
1 = Disable the port overload and UV/OV timers
PRESET
STATE
0
0
Common
Class Limit2, Class Limit1
0,0 = 160-mA classification current limit
0,1 = 80-mA classification current limit
1,0 = 40-mA classification current limit
1,1 = 20-mA classification current limit
0
D5
Class limit 1
D4
LED blink enablle
Port
0 = LED is on continuous.
1 = Blink the enabled LED at a fast-blink rate. Note fast-blink rate (in
milliseconds) equivalent to 0.00013 x TPS2383 clock (CLK). This could be
as fast as 65 ms for a 500-kHz, TPS2383 clock.
0
D3
LED low-side enable
Port
1 = Enable the low-side FET and drive the LED pin low
0
D2
LED high-side enable
Port
1 = Enable the high-side FET and drive the LED pin to V5
0
Enable modes
Port
00 = Port OFF or disable
01 = Discovery − classification − power on sequence
10 = Sample powered-mode current
11 = Power-down an active port
0
D1
D0
Table 8. Port Control
Write Register, Register Address = 010, Port Address = 010 (Port 3 Register)
BIT
FUNCTION
D7
Port fault disable
D6
Spare
RANK
Port
STATE
0 = Normal mode
1 = Disable the port overload and UV/OV timers
PRESET
STATE
0
0
Common
Class Limit2, Class Limit1
0,0 = 160-mA classification current limit
0,1 = 80-mA classification current limit
1,0 = 40-mA classification current limit
1,1 = 20-mA classification current limit
0
D5
Class limit 2
D4
LED blink enablle
Port
0 = LED is on continuous.
1 = Blink the enabled LED at a fast-blink rate. Note fast-blink rate (in
milliseconds) equivalent to 0.00013 x TPS2383 clock (CLK). This could be
as fast as 65 ms for a 500-kHz TPS2383 clock.
0
D3
LED low-side enable
Port
1 = Enable the low-side FET and drive the LED pin low
0
D2
LED high-side enable
Port
1 = Enable the high-side FET and drive the LED pin to V5
0
Port
00 = Port OFF or disable
01 = Discovery − classification − power on sequence
10 = Sample powered-mode current
11 = Power-down an active port
0
D1
Enable modes
D0
www.ti.com
19
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
Table 9. Port Control
Write Register, Register Address = 010, Port Address = 011 (Port 4 Register)
BIT
FUNCTION
D7
Port fault disable
D6
Spare
D5
A/D input select
D4
LED blink enablle
D3
D2
RANK
Port
STATE
0 = Normal mode
1 = Disable the port overload and UV/OV timers
PRESET
STATE
0
0
0 = Normal mode − select port currents
1 = Select port voltage
0
Port
0 = LED is on continuous.
1 = Blink the enabled LED at a ”fast blink” rate. Note ”fast blink” rate (in
milliseconds) equivalent to 0.00013 x TPS2383 clock (CLK). This could be
as fast as 65ms for a 500 kHz TPS2383 clock.
0
LED low-side enable
Port
1 = Enable the low-side FET and drive the LED pin low
0
LED high-side enable
Port
1 = Enable the high-side FET and drive the LED pin to V5
0
Port
00 = Port OFF or disable
01 = Discovery − classification − power on sequence
10 = Sample powered-mode current
11 = Power-down an active port
0
Common
D1
Enable modes
D0
Table 10. Port Control
Write Register, Register Address = 010, Port Address = 100−111 (Port 5 through Port 8 Register)
BIT
FUNCTION
D7
Port fault disable
D6
Spare
D5
Spare
RANK
Port
STATE
0 = Normal mode
1 = Disable the port overload and UV/OV timers
PRESET
STATE
0
0
0
D4
LED blink enablle
Port
0 = LED is on continuous.
1 = Blink the enabled LED at a ”fast blink” rate. Note ”fast blink” rate (in
milliseconds) equivalent to 0.00013 x TPS2383 clock (CLK). This could be
as fast as 65ms for a 500 kHz TPS2383 clock.
D3
LED low-side enable
Port
1 = Enable the low-side FET and drive the LED pin low
0
D2
LED high-side enable
Port
1 = Enable the high-side FET and drive the LED pin to V5
0
Port
00 = Port OFF or disable
01 = Discovery − classification − power on sequence
10 = Sample powered-mode current
11 = Power-down an active port
0
D1
Enable modes
D0
0
Run Current/Voltage
During power delivery, the average value of the port running current or voltage is available from the read register.
The slope type converter used produces 12 bits of input offset corrected conversion with a typical integration
or averaging period of approximately one line cycle (16 ms). The actual averaging period is set by the CLK
frequency and is equivalent to 8192 periods of that frequency. The lower eight bits of this conversion are
available at the port register 011. The remaining upper four bits and A/D status bits are available at register 100.
The converter span is 4096 bits. The A/D conversion port displays the real time counter output and holds the
final static value at the completion of conversion
20
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
FUNCTIONAL DESCRIPTION
Table 11. Common Analog to Digital Conversion Port
Read Register, Register Address = 011
BIT
FUNCTION
D7
A2D bit 7
D6
A2D bit 6
D5
A2D bit 5
D4
A2D bit 4
D3
A2D bit 3
D2
A2D bit 2
D1
A2D bit 1
D0
A2D bit 0
STATE
PRESET
STATE
A/D lower bits
Table 12. Common Analog to Digital Conversion Port
Read Register, Register Address = 100
BIT
FUNCTION
STATE
D7
A2D zero cross
0 = CINT is above zero threshold
1 = CINT is below zero threshold
D6
A2D overflow
1 = A2D overflow detected
D5
Reserved for test
D4
Reserved for test
D3
A2D bit 11
D2
A2D bit 10
D1
A2D bit 9
D0
A2D bit 8
PRESET
STATE
A/D upper bits
Table 13. Chip Identification/Revision
Read Register, Register Address = 101
BIT
FUNCTION
D7
Rev ID MSB
D6
Rev ID
D5
Rev ID LSB
D4
Device ID MSB
D3
Device ID
D2
Device ID
D1
Device ID
D0
Device ID LSB
STATE
Internally hardwired (function of revision)
PRESET
STATE
0
1
www.ti.com
21
SLUS559B − APRIL 2003 − REVISED JULY 2004
APPLICATION INFORMATION
SWITCH/HUB
POWERED DTR
RJ−45
CT Choke
1
TX
2
CT Choke
3
RX
6
4
RJ−45
w/grn
w/grn
grn
2
w/org
org
org
blue
blue
CT Choke
RX
grn
w/org
1
3
CT Choke
TX
6
4
PPTC Fuse
TPS2383
PSE P
(1 Port)
N
V48
GND
5
7
G
8
RS
568A
w/blue
w/blue
w/brn
w/brn
brn
brn
5
7
PD
Signature
8
PD
DC/DC
Supply
568A
Up to 350 feet of category 5 cable
RG
Optically Coupled
I2C Serial Bus
+
48−V
Supply
−
MSP430
Controller
UDG−03061
Figure 2. System Block Diagram
22
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
APPLICATION INFORMATION
V5
SCL BUS
I2C BUS
SDA BUS
INTB BUS
CT
0.1 µF 120 pF
CR
0.1 µF
+48V BUS
NEG
POS
4.22 kΩ
V5
64
63
62
61
60
59
58
57
56
55
54
53
52
51
50
49
A1
A2
A3
A4
A5
INTB
SDA_I
SDA_O
DG
SCL_I
VL
EN
CT
AG2
CR
PORB
46
4
1RS
8RS
45
5
2RS
7RS
44
6
2G
7G
43
7
2N
7N
42
8
2P
7P
41
9
3P
6P
40
10 3N
6N
39
11 3G
6G
38
12 3RS
6RS
37
13 4RS
5RS
36
14 4G
5G
35
15 4N
5N
34
16 4P
5P
33
L1
L2
L3
L4
L5
L6
L7
L8
V5
AG1
V48
V10
RG
CINT
RT
RD
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
TPS2383
OCTAL PSE CONTROLLER
17
RJ45−1/
RJ45−2/
Xformer
Xformer
RJ45−3/
RJ45−4/
Xformer
8G
RJ45−5/
1G
Xformer
0.47 µF
47
3
RJ45−6/
0.5 Ω
IRFD110 0.5 Ω
48
8N
Xformer
0.47 µF
8P
1N
RJ45−7/
0.47 µF
1P
2
Xformer
0.5 Ω
IRFD110 0.5 Ω
1
RJ45−8/
0.47 µF
Xformer
Xformer
PORB
0.47 µF
0.5 Ω
0.5 Ω IRFD110
0.47 µF
0.47 µF
0.5 Ω
0.5 Ω IRFD110
0.47 µF
V5
CD
0.68 µF
0.1 µF
0.1 µF
0.1 µF
CINT
0.033 µF
RD
665 Ω
RT
120 kΩ
Kelvin RS Sense
V5
100 Ω
Green
Green
Green
Green
Green
Green
Green
Green
100 Ω
Red
100 Ω Red
100 Ω Red
100 Ω
Red
100 Ω
Red
100 Ω Red
100 Ω
Red
100 Ω
Red
PORT 1
PORT 2
PORT 3
PORT 4
PORT 5
PORT 6
PORT 7
PORT 8
STATUS
STATUS
STATUS
STATUS
STATUS
STATUS
STATUS
STATUS
UDG−03064
* All fuses are Self Resetting BBR550, Raychem
Figure 3. Typical Eight-Port Application
www.ti.com
23
SLUS559B − APRIL 2003 − REVISED JULY 2004
APPLICATION INFORMATION
Controller VDD
SCL Bus
SDA Bus
Controller GND
CMOS
HEX
Inverter
CMOS
HEX
Inverter
OP1
PS9701
220 Ω
OP2
PS9701
OP3
PS9701
220 Ω
Isolated Ethernet
Power System
330 Ω
1 kΩ
1 kΩ
Floating +5V
Floating GND
60
59
58
57
56
55
SCL_I
61
DG
62
SDA_O
63
SDA_I
64
54
53
52
51
50
49
TPS2383
UDG−03065
Figure 4. Using Optoisolators for I2C Bus/System Ground Isolation
24
www.ti.com
SLUS559B − APRIL 2003 − REVISED JULY 2004
APPLICATION INFORMATION
0.1 µF
49
PORB
50
CR
51
AG2
52
CT
53
EN
54
VL
58
SDA_I
55
59
INTB
SCL_I
60
A5
56
61
A4
DG
62
A3
57
63
A2
SDA_O
64
A1
4.22 kΩ
L2
L3
L4
L5
L6
L7
L8
V5
AG1
V48
V10
RG
CINT
RT
18
19
20
21
22
23
24
25
26
27
28
29
30
31
+48 V Bus
Positive
47 Ω
+5 V
+
+48 V Bus
Negative
32
L1
17
No LED Drive
RD
TPS2383
0.1 µF
0.1 µF
80 V
NMOS
5 V LDO
OUT
GND
IN
+8 V
0.1 µF
10 µF
Tantalum
UDG−03066
Figure 5. V5 and VL Generation from Single +48-V Supply
www.ti.com
25
MECHANICAL DATA
MTQF008A – JANUARY 1995 – REVISED DECEMBER 1996
PM (S-PQFP-G64)
PLASTIC QUAD FLATPACK
0,27
0,17
0,50
0,08 M
33
48
49
32
64
17
0,13 NOM
1
16
7,50 TYP
Gage Plane
10,20
SQ
9,80
12,20
SQ
11,80
0,25
0,05 MIN
0°– 7°
0,75
0,45
1,45
1,35
Seating Plane
0,08
1,60 MAX
4040152 / C 11/96
NOTES: A.
B.
C.
D.
All linear dimensions are in millimeters.
This drawing is subject to change without notice.
Falls within JEDEC MS-026
May also be thermally enhanced plastic with leads connected to the die pads.
POST OFFICE BOX 655303
• DALLAS, TEXAS 75265
1
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications,
enhancements, improvements, and other changes to its products and services at any time and to discontinue
any product or service without notice. Customers should obtain the latest relevant information before placing
orders and should verify that such information is current and complete. All products are sold subject to TI’s terms
and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in
accordance with TI’s standard warranty. Testing and other quality control techniques are used to the extent TI
deems necessary to support this warranty. Except where mandated by government requirements, testing of all
parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for
their products and applications using TI components. To minimize the risks associated with customer products
and applications, customers should provide adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right,
copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or process
in which TI products or services are used. Information published by TI regarding third-party products or services
does not constitute a license from TI to use such products or services or a warranty or endorsement thereof.
Use of such information may require a license from a third party under the patents or other intellectual property
of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of 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.
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.
Following are URLs where you can obtain information on other Texas Instruments products and application
solutions:
Products
Applications
Amplifiers
amplifier.ti.com
Audio
www.ti.com/audio
Data Converters
dataconverter.ti.com
Automotive
www.ti.com/automotive
DSP
dsp.ti.com
Broadband
www.ti.com/broadband
Interface
interface.ti.com
Digital Control
www.ti.com/digitalcontrol
Logic
logic.ti.com
Military
www.ti.com/military
Power Mgmt
power.ti.com
Optical Networking
www.ti.com/opticalnetwork
Microcontrollers
microcontroller.ti.com
Security
www.ti.com/security
Telephony
www.ti.com/telephony
Video & Imaging
www.ti.com/video
Wireless
www.ti.com/wireless
Mailing Address:
Texas Instruments
Post Office Box 655303 Dallas, Texas 75265
Copyright  2004, Texas Instruments Incorporated