Echelon, LNS, LONMARK,LonTalk,LONWORKS,Neuron,
NodeBuilder, 3120, 3150, and the Echelon logo are
trademarks of Echelon Corporation registered in the United
States and other countries.
Other brand and product names are trademarks or
registered trademarks of their respective holders.
Smart Transceivers, Neuron Chips, and other OEM Products
were not designed for use in equipment or systems, which
involve danger to human health or safety, or a risk of
property damage and Echelon assumes no responsibility or
liability for use of the Smart Transceivers or Neuron Chips in
such applications.
Parts manufactured by vendors other than Echelon and
referenced in this document have been described for
illustrative purposes only, and may not have been tested
by Echelon. It is the responsibility of the customer to
determine the suitability of these parts for each
application.
ECHELON MAKES AND YOU RECEIVE NO WARRANTIES OR
CONDITIONS, EXPRESS, IMPLIED, STATUTORY OR IN ANY
COMMUNICATION WITH YOU, AND ECHELON SPECIFICALLY
DISCLAIMS ANY IMPLIED WARRANTY OF MERCHANTABILITY
OR FITNESS FOR A PARTICULAR PURP O SE .
No part of this publication may be reproduced, stored in a
retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording, or
otherwise, without the prior written permission of Echelon
Corporation.
A LONWORKS® router connects two communications channels within a
ONWORKS network, and routes LonTalk
L
ONWORKS router supports the installation of small or large networks, with
L
dozens to thousands of nodes.
This document describes how to design and develop a L
the Echelon Router 5000 chip, the Echelon FT Router 5000 chip, or the Echelon
RTR-10 Router Core Module.
This document does not describe any of the following Echelon prepackaged router
products: MPR-50 Multi-Port Router, i.LON
3000 Power Line/RF Bridge, or LonPoint Router.
Audience
This guide provides user instructions for users of LONWORKS routers. It also
provides information for developers who plan to develop a L
based on either the Router 5000 chip or the FT Router 5000 chip, or who plan to
integrate the RTR-10 router into embedded or standalone routers.
Related Documen tation
The following manuals are available from the Echelon Web site
(www.echelon.com
applications for Neuron® Chip or Smart Transceiver devices:
) and provide additional information that can help you develop
®
messages between them. Using a
ONWORKS router based on
®
600 LONWORKS/IP Server, CRD
ONWORKS router
•Connecting a Neuron 5000 Processor to an External Transceiver
Engineering Bulletin (005-0202-01D). This bulletin describes how to
connect a Neuron 5000 Processor’s communications port to external
transceivers for TP/XF-1250 channels or for EIA-485 networks, using an
external transceiver circuit. It also describes how to connect a Neuron
5000 Processor to a link-power TP/FT-10 channel using a L
LPT-11 Link Power Transceiver.
•FT 3120 / FT 3150 Smart Transceiver Data Book (005-0139-01D). This
manual provides detailed technical specifications on the electrical
interfaces, mechanical interfaces, and operating environment
characteristics for the FT 3120
•Introduction to the LONWORKS Platform (078-0391-01B). This manual
provides an introduction to the ISO/IEC 14908 (ANSI/CEA-709.1 and
EN14908) Control Network Protocol, and provides a high-level
introduction to L
are used for developing, installing, operating, and maintaining them.
•Junction Box and Wiring Guidelines for Twisted Pair LonWorks Networks
(005-0023-01P). This bulletin identifies the different types of junction
boxes and interconnections that can be used in twisted pair L
networks in building and industrial control applications.
ONWORKS networks and the tools and components that
®
and FT 3150® Smart Transceivers.
ONWORKS
ONWORKS
•LonMark
®
Application Layer Interoperability Guidelines. This manual
describes design guidelines for developing applications for open
interoperable L
Web site, www.lonmark.org
ONWORKS devices, and is available from the LonMark
01G). This manual provides specifications and user instructions for the
FTT-10A Free Topology Transceiver.
•LonWorks LPT-11 Link Power Transceiver User's Guide (078-0198-01A).
This manual provides technical specifications on the electrical and
mechanical interfaces and operating characteristics for the LPT-11 Link
Power Transceiver.
01C). This manual provides detailed specifications on the electrical and
mechanical interfaces and operating environment characteristics for the
TPT/XF-78 and TPT/XF-1250 transceiver modules.
®
•NodeBuilder
how to develop a L
FX User’s Guide (078-0405-01A). This manual describes
ONWORKS device using the NodeBuilder tool.
•PL 3120 / PL 3150 / PL 3170 Power Line Smart Transceiver Data Book
(005-0193-01C). This manual provides detailed technical specifications
on the electrical interfaces, mechanical interfaces, and operating
environment characteristics for the PL 3120, PL 3150, and PL 3170™
Smart Transceivers.
•Series 5000 Chip Data Book (005-0199-01C). This manual provides
detailed specifications on the electrical interfaces, mechanical interfaces,
and operating environment characteristics for the FT 5000 Smart
Transceiver and Neuron 5000 Processor.
All of the Echelon documentation is available in Adobe
PDF files, you must have a current version of the Adobe Reader
download from Adobe at: get.adobe.com/reader
For information about previous generation Neuron Chips, see one of the Neuron
Chip Data Books: Motorola
Neuron Chip TMPN3150/3120, or Cypress™ Neuron Chip Technical Reference
Manual.
Getting Support
You can get technical support for any of Echelon’s current product offerings by
contacting Echelon Support: www.echelon.com/support
You can also search the Echelon Knowledge Base for known product issues:
www.echelon.com/support/kb/search.asp
of information about Echelon products and technologies, including technical
articles that range from “How to” articles that describe how to complete a specific
task to “Bug” articles that document known issues with Echelon products.
PDF format. To view the
, which you can
.
®
LONWORKS Technology Device Data, Toshiba
.
. The Knowledge Base contains a wealth
iv
FCC Notice
The RTR-10 Router Core Module is designed to comply with the limits for a Class
B digital device, pursuant to Part 15 of the FCC Rules. The Router 5000 chip is
designed to comply with FCC Part 15 Subpart B and EN 55022 Level B.
These limits are designed to provide reasonable protection against harmful
interference when the equipment is operated in a commercial environment. This
equipment generates, uses, and can radiate radio frequency energy and, if not
installed and used in accordance with the instruction manual, may cause harmful
interference to radio communications. However, there is no guarantee that
interference will not occur in a particular installation.
Caution: Changes or modifications not covered in this manual must be
approved in writing by the manufacturer’s Regulatory Engineering
department. Changes or modifications made without written approval
may void the user’s authority to operate this equipment.
VDE Notice
The RTR-10 Router Core Module product is designed to comply with VDE 0871
Level B as a peripheral device. To ensure continued compliance, this product
should only be used in conjunction with other compliant devices.
Canadian DoC Notice
The RTR-10 Router Core Module digital apparatus does not exceed the Class A
limits for radio noise emissions from digital apparatus set out in the Radio
Interference Regulations of the Canadian Department of Communications.
L’ appareil RTR-10 Router Core Module numérique n’émet pas de bruits
radioélectriques dépassant les limites applicables aux appareils numériques de la
classe A prescrites dans le règlement sur la brouillage radioélectrique édicté par
le Ministère des Communications du Canada.
Table of Contents
Welcome ......................................................................................................... iii
Audience ........................................................................................................ iii
Related Documentation ................................................................................ iii
Getting Support ............................................................................................. iv
FCC Notice ..................................................................................................... v
VDE Notice ..................................................................................................... v
Canadian DoC Notice ..................................................................................... v
Introduction to LONWORKS Routers .......................................................... 1
Determining RTR-10 Firmware Version ................................................. 117
Router Firmware Version .......................................................................... 118
1
Introduction to LONWORKS Routers
This chapter describes the router theory of operation, including router
types, LonTalk protocol support for routers, and router use of message
buffers.
ONWORKS Router User’s Guide 1
L
Introduction
LONW
ORKS
Device 1
LONW
ORKS
Device
2
L
ONWORKS
Device 3
LONW
ORKS
Device 4
TP/XF-1250 Backbone Channel
TP/FT-10 ChannelTP/XF-78 Channel
TP/XF-1250
to
TP/FT-10
Router
TP/XF-1250
to
TP/XF-78
Router
In general terms, a router is a device that forwards data packets between
communications networks. The router connects to the data lines from each
network, and reads address information in each data packet to determine the
packet’s destination.
ONWORKS router connects two communications channels within a LONWORKS
A L
network, and routes LonTalk messages between them. Using a L
router supports the installation of small or large networks, with dozens to
thousands of nodes.
Figure 1 shows a typical router installation, with a free topology channel and a
78 kbps bus topology channel connected to a 1.25 Mbps backbone twisted pair
channel. Because the network includes the routers, applications on each of the
ONWORKS devices can communicate with each other transparently, as if they
L
were installed on a common channel.
ONWORKS
A single router can connect two channels, or multiple routers (called redundant
routers) can connect the same pair of channels. Redundant routers provide fault
tolerance by providing more than one routing path from one channel to another.
They are also required when not all devices on a given channel are able to hear
one another (referred to as an “ear shot problem”), for example on a radio
frequency channel. For a router to function as a redundant router, the router
must be configured to be a Configured router (see Router Types).
ONWORKS routers are used to:
L
2 Introduction to LONWORKS Routers
Figure 1. Sample Router Installation
•Extend the limits of a single channel. You can use a router to add a
channel to a L
ONWORKS network to support additional devices or to
extend the maximum channel length. You can add multiple routers,
depending on the capacity or distance needed.
•Interface different communications media, or bit rates, in a L
network. For example, you might want to trade data rate for distance on
portions of the network, or to use a 1.25 Mbps backbone twisted pair
channel to connect several 78 kbps free topology and link power channels.
ONWORKS
Alternatively, you might want to use power line for a portion of the
network where the devices are subject to frequent physical relocation, or
if cable installation is difficult. For each of these cases, you use a router
to connect the dissimilar L
ONWORKS channels.
•Enhance the reliability of the L
ONWORKS network. The two channels that
connect to a router are logically isolated, so a failure on one channel does
not affect the other channel. For example, in an industrial control
network, isolation among connected cells might be desirable to prevent a
failure in a single cell from bringing down multiple cells. You can achieve
this goal by dedicating channels to individual cells and isolating them
from one another with routers.
•Improve overall network performance. You can use routers to isolate
traffic within subsystems. For example, in a cluster of industrial cells,
most of the communications might be between devices within cells rather
than across cells. Using intelligent routers across cells avoids forwarding
messages addressed to devices within specific cells, thus increasing the
capacity and decreasing the response time of the overall network.
The use of routers across channels is transparent to the application programs
within devices. Thus, you can develop applications without needing to know the
workings of the routers or even if the device’s channel will use a router. You only
need to consider routers when determining the network image of a device. When
you move a device from one channel to another, you need only change the
network image. Use a network management tool, such as OpenLNS CT, to
manage network images.
LONWORKS Router Products
Echelon provides the following router products:
•MPR-50 Multi-Port Router (Model: 42150)
Five-channel (one TP/XF-1250 channel and four TP/FT-10 channels)
ONWORKS router. The MPR-50 can be used to connect two, three, or
L
four TP/FT-10 channels together, or it can be used to connect these
TP/FT-10 channels to a high-speed TP/XF-1250 backbone.
•i.LON 600 L
ONWORKS/IP Server (Model: 7260x)
An EIA-852 compliant LonTalk-to-IP router. The i.LON 600 provides
secure Internet access to L
ONWORKS devices and transforms the Internet
(or other IP-based network) into a pathway for L
information.
•CRD 3000 Power Line/RF Bridge (Model: 76520R)
A Power Line (PL) to RF communications device, designed primarily for
intelligent L
ONWORKS street lighting networks.
•LonPoint Router (Model: 4210x)
A two-channel router for TP/FT-10, TP/XF-78, or TP/XF-1250 L
channels. Three models are available for various network connection
combinations.
ONWORKS control
ONWORKS
ONWORKS Router User’s Guide 3
L
•RTR-10 Router Core Module (Model: 61000R)
A compact module used by OEMs to build L
ONWORKS routers. The RTR-
10 consists of the core electronics and firmware needed to implement a
router.
•Router 5000 (Model: 14315R)
A semiconductor product used by OEMs to build half-routers or full
routers for various L
ONWORKS channel types. The Router 5000 includes
the firmware required to implement a half-router.
•FT Router 5000 (Model: 14285R)
A semiconductor product used by OEMs to build half-routers or full
routers for L
ONWORKS Free Topology (FT) channels. The FT Router 5000
includes the firmware required to implement a half-router.
Packaged routers eliminate the need to build hardware and obtain the necessary
electrical interference and safety certifications. Thus, they allow direct, off-the-
shelf integration into the user’s L
ONWORKS network. This manual does not
describe how to use the Echelon prepackaged router products. See the Echelon
router Web page (www.echelon.com/products/routers
) for information about the
pre-packaged Echelon router products.
This manual describes those Echelon router products that allow OEMs to design
and build their own custom routers for L
ONWORKS channels: the RTR-10 Router
Core Module, the Router 5000 chips, and the FT Router 5000 chips.
RTR-10 Overview
The RTR-10 Router Core Module is a compact module used by OEMs to design
and build L
channels and route LonTalk messages between them. They support installation
of networks with dozens to thousands of devices.
The RTR-10 Module consists of the core electronics and firmware required to
implement a router. Its compact single inline module (SIM) form factor
minimizes the board space required to implement a router. Vertical SIM sockets
are available to minimize board space; right-angle SIM sockets are also available
to minimize component height.
A complete router, using an RTR-10 module, consists of the module, two
transceivers, and a motherboard to connect the RTR-10 router to the two
transceivers, as shown in Figure 2.
ONWORKS routers. A LONWORKS router connects two communications
4 Introduction to LONWORKS Routers
RTR-10 Router Core Module
Side A
Transceiver
Side B
Transceiver
Side A
Network
Connector
Side B
Network
Connector
Power Supply
Service
Button and
LEDs
Figure 2. Block Diagram of a LonWorks Router Based on the RTR-10
As the figure shows, an RTR-10 router and two transceiver modules (one to
handle each of two channels connected by the router) can be mounted on a
motherboard, along with a single power supply and two network connectors.
This sub-assembly constitutes a L
ONWORKS router. It can be packaged in an
enclosure to meet unique form factor and environmental requirements.
Depending on the application, the package could contain a single router sub-
assembly, or could include other application-specific hardware. Multiple routers
can be packaged together for some applications, such as a backbone connecting
multiple channels.
The RTR-10 router comes preconfigured with many common L
ONWORKS
transceiver parameters. Two sets of five transceiver identification (XID) pins on
the RTR-10 router select the appropriate transceiver type for each side. The
transceiver ID inputs eliminate a manufacturing step by automatically
configuring the RTR-10 router for most transceivers. A special transceiver ID is
reserved for programming any custom type.
One side of the RTR-10 router has a fixed input clock rate of 10 MHz. This side
can be used with transceivers running at interface bit rates from 9.8 kbps to 1.25
Mbps. The second side of the RTR-10 router can be tied to the 10 MHz output of
the first side, requiring no external components for interface bit rates from 9.8
kbps to 1.25 Mbps. Alternatively, the 10 MHz output can be divided to a lower
frequency with external hardware and used as the input clock for the second side
to support transceivers running at bit rates as low as 610 bps.
Any pair of channel types can be connected by a router by selecting the
appropriate pair of transceivers. The RTR-10 router is compatible with all
ONWORKS Router User’s Guide 5
L
LONWORKS transceivers, including standard transceivers for free topology, link
power, twisted pair, and power line. Using multiple communications media can
minimize installation costs and increase system performance by allowing easily
installed media, such as power line or link power, to be combined with high-
performance media such as TP/XF-1250 twisted pair.
Router 5000 and FT Router 5000 Overview
The Router 5000 chip and the FT Router 5000 chip (generically referred to as
Series 5000 router chips) are Echelon semiconductor products, based on the
Echelon Neuron 5000 Core, that are used to build half-routers and full routers for
ONWORKS channels. A LONWORKS router connects two communications
L
channels and route LonTalk messages between them. They support installation
of networks with dozens to thousands of devices.
Both the Router 5000 and FT Router 5000 include the Router firmware required
to implement a half-router. Each chip’s compact form factor minimizes the space
required to develop a half-router. You can implement two half-routers to develop
a full router for the same, or different, external transceiver types.
The FT Router 5000 integrates the high performance Neuron 5000 Core with a
free topology (FT) twisted pair transceiver. Thus, the FT Router 5000 supports
ONWORKS TP/FT-10 channel type.
the L
Table 1 lists commonly used channel and transceiver types for Router 5000-
based router halves; see Developing a Router with the Router 5000 Chip for
additional information about connecting a Router 5000 chip to these external
transceiver types. These external transceivers can run at interface bit rates from
9.8 kbps to 1.25 Mbps.
Table 1. Common Channel and Transceiver Types for a Router 5000
Channel Type Transceiver for Half Router Notes
TP/FT-10 Echelon FTT-10A Free Topology
Transceiver (Model 50051)
TP-RS485 Any EIA-485 transceiver Can use any standard 3.3V or 5V
TP/XF-78 Echelon TPT Twisted Pair
Transceiver Module (Model 50010)
TP/XF-1250 Echelon TPT Twisted Pair
Transceiver Module (Model 50020)
Connection to the Router 5000 is
similar to the connection to a
Neuron Chip.
Use an FT Router 5000 for a
standard (non-link-powered)
TP/FT-10 channel or for a locally
powered device on a link-power
TP/FT-10 channel.
EIA-485 transceiver.
Add differential driver and
differential comparator circuits
(contact Echelon Support).
Add differential driver and
differential comparator circuits.
6 Introduction to LONWORKS Routers
Channel Type Transceiver for Half Router Notes
Side A
Transceiver
Side B
Transceiver
Side A
Network
Connector
Side B
Network
Connector
Power Supply
Service
Buttons and
LEDs
Router 5000
Half Router
Router 5000
Half Router
Side A
Network
Connector
Side B
Network
Connector
Power Supply
Service
Buttons and
LEDs
FT Router
5000
(with FT-X3)
Half Router
FT Router
5000
(with FT-X3)
Half Router
LonWorks Router Based on the
Router 5000
LonWorks Router Based on the
FT Router 5000
Link-power Echelon LPT-11 Link Power
Transceiver (Model 50040)
Echelon provides special licensing for other transceiver types, such as a Power
Line Smart Transceiver; contact Echelon Support for additional information.
A complete router using the Router 5000 consists of two Router 5000 half routers,
two transceivers, and a motherboard to connect the two half routers. A complete
router using the FT Router 5000 consists of two FT Router 5000 half routers
(each with its own FT-X3 Communications Transformer), and a motherboard to
connect the two half routers. Figure 3 shows block diagrams of both types of
Series 5000 based routers.
Add linear regulator and TX buffer
circuit.
Figure 3. Block Diagrams of L
ONWORKS Routers Based on a Series 5000 Router Chip
In the left side of the figure, two Router 5000 half routers and two transceiver
ONWORKS Router User’s Guide 7
L
modules, one to handle each of two channels connected by the router, can be
mounted on a motherboard, along with a single power supply and two network
connectors. This sub-assembly constitutes a L
of the figure, two FT Router 5000 half routers, one to handle each of two channels
connected by the router, can be mounted on a motherboard, along with a single
ONWORKS router. In the right side
power supply and two network connectors. This sub-assembly constitutes a
LONWORKS router. Although not shown in the figure, you can place a Router
5000 half router (with its transceiver module) and FT Router 5000 half router on
the same motherboard to create a L
ONWORKS router.
A complete L
form factor and environmental requirements. Depending on the application, the
package could contain a single router sub-assembly, or could include other
application-specific hardware. Multiple routers can be packaged together for
some applications, such as a backbone connecting multiple channels.
Unlike an RTR-10 router, you store the L
each Router 5000 or FT Router 5000 half router in external EEPROM, thus
allowing you to define the appropriate transceiver type for each side.
ONWORKS router can be packaged in an enclosure to meet unique
ONWORKS transceiver parameters for
Comparing the RTR-10, Router 5000, and FT Router
5000
Table 2 lists some of the major characteristics of the three router products
described in this manual.
Table 2. Comparing the RTR-10, Router 5000, and FT Router 5000
Parameter RTR-10 Router 5000 FT Router 5000
Package Integrated module with
both A side and B side
Voltage 5 V 3.3 V 3.3 V
7mm x 7mm QFN chip 7mm x 7mm QFN chip
Clock rate 10 MHz A side
≤10 MHz B side
Router Buffers 1408 bytes 25 KB 25 KB
Transceivers Any TP/FT-10
The main difference between the Router 5000 and the FT Router 5000 is their
transceiver support. The FT Router 5000 supports only the TP/FT-10 channel,
but because the transceiver is integrated into the chip, you do not need to use an
FTT-10A Free Topology Transceiver (and associated circuitry to accommodate the
different voltage standards between the half router and transceiver).
5, 10, 20, or 40 MHz
system clock
(10 MHz crystal)
TP-RS485
TP/XF-78
TP/XF-1250
LPT-11 Link-power
Other (contact Echelon
Support)
5, 10, 20, or 40 MHz
system clock
(10 MHz crystal)
TP/FT-10
(requires an FT-X3
Communications
Transformer)
8 Introduction to LONWORKS Routers
Router Types
A LONWORKS router can use one of four routing algorithms: configured router,
learning router, bridge, and repeater. This selection allows you to trade system
performance for ease of installation. The configured router and learning router
algorithms create intelligent routers that selectively forward messages based on
network topology. Both sides of a router must use the same routing algorithm.
The following general rules apply to all four routing algorithms:
•For a message to be forwarded, it must fit into the router’s input and
output message buffers. A free input message buffer must be available.
•For a message to be forwarded, it must have a valid cyclic redundancy
check (CRC) code.
•Priority messages are forwarded as priority messages, but with the
priority level of the transmitting side rather than the priority level of the
originator of the message. If the transmitting side has not been installed
with a priority value, then priority messages are not forwarded in a
priority slot. The priority message is still flagged as a priority message,
so that if it passes through a second router that is installed with a
priority level, the second router transmits the message in a priority slot.
Repeater
A Repeater is a router that forwards all messages in both directions, regardless of
the message’s destination or domain. That is, a repeater forwards all valid
messages (that is, messages with a valid CRC code) to the other channel.
A Permanent Repeater behaves similarly, but its type cannot be changed after
creation.
Bridge
A Bridge is a router that forwards all messages received on either of the router’s
domains, regardless of the message’s destination. That is, a bridge forwards
packets received on one channel to the other channel, if the packet is sent on a
domain to which the bridge belongs. Use a bridge to span domains. In a single
domain network, a bridge functions essentially the same as a repeater.
A Permanent Bridge behaves similarly, but its type cannot be changed after
creation.
Configured Rout er
A Configured Router determines which packets to forward based on internal
routing tables. A configured router forwards only those messages which are
received on either of the router’s domains and which meet the forwarding rules
shown in Figure 4 and Figure 5. Configured routers maintain their routing
tables in non-volatile memory, and thus retain them after a reset. These tables
control forwarding of subnet and group-addressed messages, and are managed by
a network management tool.
ONWORKS Router User’s Guide 9
L
A forwarding table is used for each domain on each side of the router. Each
forwarding table contains a forwarding flag for each of the 255 subnets and 255
groups in a domain. As shown in Figure 4 and Figure 5, these flags determine
whether or not a message should be forwarded or dropped based on the
destination subnet or group address of the message.
A network management tool initializes the forwarding tables using the network
management messages described in Chapter 7, Network Management Messages,.
By configuring the routing tables based on network topology, a network
management tool can optimize network performance and make the most efficient
use of available bandwidth. Configured routers should be used for looping
topologies; see Loop Topology.
For a L
volatile memory (typically EEPROM) and one in RAM. The non-volatile table is
copied to the RAM table when the router is initially powered-up, after a reset,
and when the router receives the Set Router Mode command with the Initialize
Routing Table option. The RAM table is used for all forwarding decisions.
Several of the operations in shown in Figure 4 and Figure 5 help prevent
message loops for service-pin messages. Service-pin messages require special
handling because they are broadcast to all nodes on the zero-length domain, and
have a source subnet ID of zero. When a router receives a service-pin message
with a source subnet ID of zero, the router modifies the source subnet field of the
message to be the router’s subnet on the receiving side. If the receiving side is
installed in two domains, two service-pin messages are forwarded, one for each
domain. Thus, the router can drop the service-pin message if a loop causes the
message to be received again on the same side.
ONWORKS router, there are two sets of forwarding tables, one in non-
Learning Router
A Learning Router, like a configured router, determines which packets to forward
based on internal routing tables. A learning router forwards only those messages
which are received on either of the router’s domains and which meet the
forwarding rules shown in Figure 4 and Figure 5. A learning router always
forwards all group-addressed messages. Learning routers maintain their routing
tables in non-volatile memory, and thus retain them after a reset. These tables
control forwarding of subnet and group-addressed messages, and are updated
automatically by the router firmware, rather than their being configured by a
network management tool. The group forwarding tables are configured to always
forward (flood) all messages with group destination addresses.
When a router receives a packet with a destination address using a subnet ID, it
uses the subnet ID to determine whether to forward the packet. Learning
routers learn network topology by examining the source subnet of all messages
received by the router. Whenever a learning router receives a packet from one of
its channels, it uses the source subnet ID to learn the network topology. It sets
the corresponding routing table entries to indicate that the subnet in question is
to be found in the direction from which the packet was received. Because subnets
cannot span two channels connected to an intelligent router, the router can learn
which side a subnet is on whenever that subnet ID appears in the source address.
The subnet forwarding tables are initially configured to forward all messages
with subnet destination addresses. Each time a new subnet ID is observed in the
source address field of a message, its corresponding flag is cleared (that is,
forwarding is disabled) in the subnet forwarding table. The forwarding flag for
10 Introduction to LONWORKS Routers
the destination address is then checked to determine whether the message
Router receives
message packet
Is message addressed
to either of the router’s
domains?
Is message domain
length zero,
source
subnet zero
, addressed
as broadcast?
Drop packet
Is message source
subnet zero
?
Is message domain
length zero
,
addressed
as broadcast?
Clear subnet fwd flag
Set message source subnet
=
router subnet
Substitute one or both
domains from the router
,
source subnet = our
subnet, dest subnet zero
Forward one or two
packets
No
YesYes
Yes
No
Yes
NoNo
Page 2
*
Applies to
learning
router only
should be forwarded or dropped. The forwarding flags are all cleared whenever
the router is reset, so the learning process restarts after a reset.
The forwarding flag for a given subnet should never be cleared on both sides of a
router. However, the flag can be cleared on both sides if a device is moved from
one side of a router to the other side. For example, if subnet 1 is located on side
A of a router, the router will learn subnet 1’s location as soon as it receives a
message generated by any device in subnet 1. If any subnet 1 device is moved to
side B without reinstalling it, the router will learn that subnet 1 is also on side B,
and will stop forwarding subnet 1 messages to side A. The router detects this
error and logs it, as described in Chapter 7, Network Management Messages.
Figure 4. Configured and Learning Router Forwarding Rules, Part 1
ONWORKS Router User’s Guide 11
L
Is message addressed
to group
?
Is message addressed
to subnet
/node
?
Message must be
addressed as broadcast or
48
-bit Neuron ID
Is the group fwd flag
of the dest group set to
forward
?
Drop packet
Is the subnet fwd flag
for the dest subnet set
to forward
?
Is the subnet fwd flag
for the source subnet
set to forward
?
Drop packet
Is message dest subnet
=
zero?
Yes
Yes
Yes
No
No
Yes
No
No
Yes
No
A
Yes
A
No
B
B
Page
1
Forward packet
*
Applies to
configured
router only
Figure 5. Configured and Learning Router Forwarding Rules, Part 2
As with configured routers, learning routers sometimes modify source addresses
for service-pin messages to help prevent message loops.
Learning routers, in general, are less efficient in using channel bandwidth
because they always forward all messages with group destination addresses.
Their advantage is simplified installation because the installation tool does not
need to know the network topology to configure the router.
Loop Topology
A looping topology is a network topology that has the potential for message loops.
A loop is a path through two or more routers that forwards a message from a
channel to itself. For example, Figure 6 shows a looping topology with two
channels and two routers. A message on channel A could be forwarded by router
12 Introduction to LONWORKS Routers
1 to channel B, then the same message could be forwarded by router 2 back to
Channel A
Router 1Router 2
Channel B
channel A, starting an endless loop of forwarded messages.
Figure 6. Looping Topology
The LonTalk protocol does not support topologies where loops can occur.
However, looping topologies can be desirable for the following reasons:
•Increased Reliability. Redundant routers can increase system reliability
by providing multiple paths between two channels.
•Support for Open Media. Open media (such as radio frequency [RF]
communications) might require redundant routers with overlapping
coverage to ensure complete coverage of an area.
You can use configured routers (see Configured Router) to support looping
topologies by configuring the routers to prevent message loops. For example, the
topology in Figure 6 can be supported if both routers are configured to forward
all messages addressed to subnets on channel B from channel A; and all
messages addressed to subnets on channel A from channel B. Any groups with
members on both channels can only be forwarded by one of the two routers.
Network management tools, such as OpenLNS CT, can automatically set up the
forwarding tables for configured redundant routers.
Power Line Router s
A looping topology can be inadvertently created when using power line (PL)
media. Passive coupling between different phases of a power line system can
cause packets transmitted on one phase to be received by devices installed on
another phase. A loop can be formed when active coupling provided by a router is
combined with passive coupling. Figure 7 shows an example looping topology
with a power line router.
ONWORKS Router User’s Guide 13
L
PL Phase A
Power Line
Router
Passive
Coupling
PL Phase B
Figure 7. A Looping Topology with One Router
Routers can be used between power line channels only if the two channels are
fully isolated. Such isolation is generally not the case between two phases on the
same circuit, but can be the case between phases on different distribution
transformers. Use an Echelon PLCA-22 Power Line Communication Analyzer to
confirm isolation between power line channels before installing power-line-topower-line routers.
LonTalk Protocol Support for Routers
The LonTalk protocol1 is designed to provide transparent routing of messages
between devices that communicate through routers. To increase the efficiency of
routers, the LonTalk protocol defines a hierarchical form of addressing using
domain, subnet, and device (node) addresses. An intelligent router operates at
the subnet level. The router determines which subnets lie on each of its two
sides, and forwards packets accordingly.
Subnets do not span intelligent routers, which allow intelligent routers to make
routing decisions based on the subnet component of a device’s logical address. To
further facilitate the addressing of multiple dispersed devices, the LonTalk
protocol defines another class of addresses using domain and group addresses.
Intelligent routers also can be configured to make routing decisions based on the
group addressing component of a message.
In general, a network management tool, such as OpenLNS CT, is responsible for
domain, subnet, node, and group address assignments.
See the ISO/IEC 14908 Control Network Protocol specification for detailed
information about the LonTalk protocol.
Message Buffers
As messages are received by a router, they are placed in an input buffer queue. By
default, this queue is limited to two message buffers to ensure that priority
messages are never enqueued behind more than one non-priority message. When
forwarded to the transmitting side of the router, priority messages have their own
1
Echelon’s implementation of the ISO/IEC 14908 Control Network Protocol is called the
LonTalk protocol. Echelon has implementations of the LonTalk protocol in several product
offerings, including the Neuron firmware, OpenLNS
network interfaces. This document refers to the ISO/IEC 14908-1 Control Network Protocol
as the “LonTalk protocol,” although other interoperable implementations exist.
14 Introduction to LONWORKS Routers
®
Server, SmartServers, and various
outgoing buffer queue. Thus, priority processing of these outgoing messages is
12
12
123
N - 1N
Message Receiving Side
Message Sending Side
Input Buffer Queue
Priority Output Buffer Queue
Normal Output Buffer Queue
Direction of Message Flow
assured because the transmitting side will send messages from the priority output
buffer queue before sending messages from the non-priority output buffer queue.
Figure 8 shows the message flow through the input and output buffer queues. This
message flow is duplicated for messages moving in the opposite direction, that is,
another set of input and output buffer queues exist for messages flowing in the
opposite direction.
Figure 8. Buffering Scheme for a L
ONWORKS Router
The size and count of the message buffers is limited by the amount of RAM on
the router.
RTR-10 Message Buffers and Transact i on Records
There are three different versions of the router firmware for the RTR-10 router.
To determine the router firmware version of your router, see Appendix B,
Determining RTR-10 Firmware Version.
Some versions of the RTR-10 routers were shipped with a reduced buffer and
transaction record capacity. This does not impact system performance. However, a
catastrophic failure of the router can occur if the buffers or transaction records on a
router with reduced buffer and transaction record capacity are reconfigured and
assume that a larger buffer and transaction record space is available. This could
occur if you are using a manufacturing test station to reconfigure routers based on
the higher router buffer and transaction record capacity. It can also occur if you
use a network installation tool that reconfigures router buffers or transaction
records when replacing a router using a non-standard configuration. The failure
condition that occurs is that the router will go into the application-less state and
will no longer function as a router. This failure mode cannot be recovered in the
field. If you are making the change with a tool based on the OpenLNS Server or
the OpenLNS Commissioning Tool (or LNS and the LonMaker Turbo Integration
Tool), in some cases, the tool will prevent an invalid configuration; however, in
other cases the tool may allow an invalid configuration and the router will fail. Use
the tables, below, to understand the correct buffer and transaction record
configurations for your RTR-10 router.
Depending on your firmware, buffers and transaction records are allocated
according to the following tables. For versions A and C, Table 3 shows the
transaction record configurations for receive and transmit transactions, and the
ONWORKS Router User’s Guide 15
L
buffer configurations for input buffers, output buffers, priority buffers, and non-
Application Buffer In
1
42
42
lication
Receive Transaction Record
3
13
39
Application Buffer In
1
42
42
priority buffers.
Table 3. RTR-10 Default Buffer and Transaction Record Configuration,
Firmware A and C
Size
Type Count
Receive Transaction Record 3 13 39
Transmit Transaction Record 2 28 56
Application Buffer Out 1 42 42
Network Buffer In 2 66 132
Network Buffer Out 15 66 990
(Bytes) Total Bytes
App
Network Buffer Out Priority 2 66 132
Total Allocated Bytes 1475
Unused 1 25 25
Total Available Bytes for Transaction
Records and Buffers
Total Available Bytes for Buffers,
including Default Transaction
Records shown in first two lines,
above
The default buffer sizes allow the router to handle packets with maximum
address overhead and data size for any network variable message and explicit
messages with up to 40 bytes of data; this is large enough for any network
management or network diagnostic message.
For RTR-10 routers with version B firmware, Table 4 shows the transaction
record configurations for receive and transmit transactions, and the buffer
configurations for input buffers, output buffers, priority buffers, and non-priority
buffers.
Table 4. RTR-10 Default Buffer and Transaction Record Configuration,
Buffer Out Priority
Firmware B
1 42 42
1500
1405
16 Introduction to LONWORKS Routers
Size
Type Count
Transmit Transaction Record 2 28 56
Application Buffer Out 1 42 42
Network Buffer In 3 66 198
Network Buffer Out 11 66 726
(Bytes) Total Bytes
Application Buffer Out Priority
1 42 42
Network Buffer Out Priority 3 66 198
Total Allocated Bytes 1343
Unused 1 65 65
Total Available Bytes for Transaction
Records and Buffers
1408
Total Available Bytes for Buffers,
including Default Transaction
Records shown in first two lines,
1313
above
You will not have a problem interchanging routers with Router Firmware
Versions A, B, or C if you are not changing the router buffer or transaction record
configuration. If you need to change the configuration, make sure that the total
number of bytes required for the buffers and transaction records does not exceed
the capacity for the version of the router firmware that you are using. For
example, to use the OpenLNS Commissioning Tool or the LonMaker Turbo
Integration Tool to safely change the buffer configuration for a router, right-click
the router shape in the OpenLNS CT or LonMaker drawing and then click
Properties on the shortcut menu. Click the Buffers tab to display and change
the buffer configuration. As you change the buffer configuration, the required
memory for each side of the router for your buffer configuration is displayed
under the Memory heading. Verify that the memory required is less than the
buffer capacity listed in Table 3 or Table 4 (as appropriate for your router
firmware version) before clicking OK or Apply. When you click OK or Apply,
the buffer configuration you selected is written to the router if you are attached
to the network and you are OnNet. If the memory required is larger than the
buffer and transaction record capacity, the router will fail to operate.
In applications that must route large explicit messages with more than 40 bytes
of data, the buffer size must be increased, and the count of nonpriority buffers
decreased. See the Neuron C Programmer’s Guide to understand how the
network buffer sizes are calculated. See Network Management Messages in this
guide for a description of how to change the size and count of buffers. You can
also use the NodeUtil Node Utility, which you can download from the Echelon
Web site. However you allocate the transaction record counts and the buffer
sizes and counts, the total memory required by the transaction records and buffer
must not exceed the total available memory size in Table 3 or Table 4 (as
appropriate for your router firmware version).
The default buffer configuration places the bulk of the buffers on the output
queues of the router. The reasoning behind this configuration is to keep buffered
packets on the output queues, after they have been processed for forwarding.
This processing includes checking for priority packets. Priority packets are
sensed and forwarded through the router’s priority output buffers, so that
priority packets are processed as quickly as possible, rather than allowing them
to be delayed behind non-priority packages in a large input queue.
There are applications, however, where the network traffic can be “bursty”,
where many packets appear on the network almost at the same time. In these
cases, the traffic bursts could cause the input queue to become full and lose
excess packets.
ONWORKS Router User’s Guide 17
L
In this case, reduce the number of lost messages by moving more of the packet
Priority Output Buffer Queue
2
255
510
buffering from the output queue to the input queue by increasing the size of the
input queue and decreasing the size of the output queue. A router with a larger
input queue can handle larger bursts of traffic, at the risk of priority messages
being queued behind a number of non-priority messages.
Router 5000 and FT Router 5000 Message Buffers
Each router side has maximum 26 623 bytes of buffer space available. Because
both the Router 5000 and FT Router 5000 have sufficient RAM available for any
router configuration, you can allocate this space with any combination of buffers,
for example, seven input buffers, two priority output buffers, and seven nonpriority buffers. You can specify any valid buffer size (see the Neuron C Programmer’s Guide for information about valid buffer sizes), but, in general,
there is no reason not to specify the maximum size of 255 bytes. Table 5 shows a
general buffer configuration.
Table 5. General Series 5000 Router Buffer Configuration
Queue Count
Input Buffer Queue 7 255 1785
Non-Priority Output Buffer Queue 7 255 1785
Total 4080
The buffer size of 255 bytes allows the router to handle packets with maximum
address overhead and data size for any network variable message or explicit
message. See Configuring a Series 5000 Half-Router for a description of how to
change the size and count of buffers. However you allocate the buffer sizes and
counts, the total memory required by the three buffer queues must not exceed 25
K bytes.
The general buffer configuration shown in Table 5 balances the buffers between
the input and output queues of the router. For systems with large bursts of
traffic, you could specify additional non-priority output buffers. Priority packets
are sensed and forwarded through the router’s priority output buffers, so that
priority packets are processed as quickly as possible, rather than allowing them
to be delayed behind non-priority packages in a large input queue.
Router Performance
A major criterion of router performance is network throughput. An optimal
router would be able to forward traffic at the wire-rate, with zero packet loss and
minimal delay. Thus, an optimal router would forward traffic from 9.8 kbps to
1.25 Mbps, depending on the router’s transceiver type.
Size(Bytes)
Total Bytes
18 Introduction to LONWORKS Routers
A real router typically does not perform at the wire-rate because of latency within
the router, including the time to receive and buffer the incoming packet at the
near side, the time to forward the packet between the halves, and the time to
buffer and transmit the packet at the far side. You should measure your router
device’s latency to determine if its design meets your system’s needs.
Example: For a 20 MHz Router 5000 device (where both halves use the Router
5000 chip), a measured data transfer rate for sending a service-pin message
between the router halves was approximately 1.2 μs per byte (or 830 kbytes/sec).
Some additional latency was seen for the time between the beginning of the
original packet transmission and the beginning of the forwarded packet
transmission.
For slower channel types, this router latency is not significant, but could become
significant for faster channel types.
The latency between router halves is relatively invariant, with respect to router
configuration, whereas overall router latency depends on the router type and
configuration. For an RTR-10 device, the maximum data transfer rate between
router halves is approximately 2.4 μs per byte (or 416 kbytes/sec). For a Router
5000 device (where both halves use the Router 5000 chip), the maximum data
transfer rate between router halves is approximately 600 ns per byte (or 1.6
Mbytes/sec, assuming a 40 MHz system clock for both halves; this rate scales
with the system clock setting). An FT Router 5000 performs similarly to the
Router 5000.
ONWORKS Router User’s Guide 19
L
Loading...
+ 99 hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.