System Overview
7
TIDUES1A – October 2019 – Revised February 2020
Copyright © 2019–2020, Texas Instruments Incorporated
EMC Compliant 10/100-Mbps Ethernet PHY Reference Design With IEEE
802.3at Type-1 (
≤
12.95 W) PoE-PD
2.2
Design Considerations
2.2.1
Ethernet Physical Layer Transceiver (PHY)
The standard 100-m limitation of Ethernet might severely limit the viable locations where system designers
and implementers want to operate a remote IP-enabled device. IP surveillance networks should be able to
withstand high frequencies and large data packet transmission with minimum data loss to support IP video
and mega-pixel transmission. The IP network cameras are becoming more compact and are always a
space-critical application. In this reference design, it is necessary to use a 10/100 Mbps Ethernet PHY that
can support a longer cable beyond the standard Ethernet distance limitation of 100 m (328 feet) in a small
form-factor at a lower cost and lower power consumption.
This reference design is based around the DP83825I device, which is an ultra-small form factor, very low
power Ethernet physical layer transceiver with integrated PMD sublayers to support 10BASE-Te,
100BASE-TX Ethernet protocols. It supports up to 150 m reach over CAT5e cable. The DP83825I device
interfaces directly to twisted pair media via an external transformer. It interfaces to the MAC layer through
RMII both in Master and Slave mode and it provides 50-MHz output clock in RMII master mode. This clock
is synchronized to MDI derived clock to reduce the jitter in the system.
2.2.2
Power-over-Ethernet (PoE)
For any IP-enabled device to function, it requires both Ethernet and power. It is often desirable and
required to install a remote IP-enabled device in a location where electrical power is not available. This is
where PoE technology supplies a solution. PoE is a mechanism for supplying power to network devices
over the same cabling used to carry network traffic. Therefore, no infrastructure upgrade is necessary.
PoE technology saves time and the cost of installing separate power cabling, AC outlets and wall warts,
as well as eliminating the need for a dedicated UPS for individual devices. PoE can enable fast installation
and deployment, and maximum reliability for current IP-based networks. As PoE changes to meet growing
technology and application requirements, it is being categorized by classes. PoE devices, on the other
hand, are classified by type depending on their power requirements. The difference between PoE “types”
and “classes” can sometimes cause confusion when talking about PoE applications and capabilities. IEEE
PoE standards provide for signaling between the PSE and PD. PSEs are devices (such as switches) that
provide power on the network cable. PDs are the devices powered by a PSE: wireless access points, IP
surveillance cameras, VoIP phones, and so forth. In this reference design, the PSE and PD must
demonstrate reliable power delivery along with data communication over the Ethernet cable.
This reference design uses the TPS23861 device as PoE-PSE and the TPS23755 device as PoE-PD. The
TPS23861 device is an easy-to-use, flexible, IEEE802.3at PSE solution. The TPS23861 device
automatically detects PDs that have a valid signature, determines power requirements according to
classification and applies power. Two-event classification is supported for Type-2 PDs. The TPS23861
device supports DC disconnection and the external FET architecture allows designers to balance size,
efficiency, and solution cost requirements. The TPS23755 device combines a PoE PD interface, a 150-V
switching power FET, and a current-mode DC/DC controller optimized for flyback topology. The high level
of integration along with primary side regulation (PSR), spread spectrum frequency dithering (SSFD), and
advanced startup makes the TPS23755 device an ideal solution for size-constrained applications. The
PoE implementation supports the IEEE 802.3at standard as a 13-W, Type 1 PD.
